diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_0_context5_2000_effbs128/eval/post_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_0_context5_2000_effbs128/eval/post_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..4f7f28f17558b7a3c90cb93b8b12961829d008e8 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_0_context5_2000_effbs128/eval/post_train/step_2000.json @@ -0,0 +1,123 @@ +{ + "per_latency": { + "flappy/latency_0": { + "latency": 0, + "num_episodes": 20, + "mean_reward": 423.83000000002437, + "mean_length": 3432.45, + "std_reward": 90.53433105734537, + "std_length": 730.3335179902397, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 29.200000000000127, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 249, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 63231, + "1": 5418 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + } + }, + "aggregate": { + "stage": "post_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 20, + "mean_reward": 423.83000000002437, + "mean_length": 3432.45, + "std_reward": 90.53433105734537, + "std_length": 730.3335179902397, + "task_count": 1, + "macro_mean_reward": 423.83000000002437, + "macro_mean_length": 3432.45, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_1000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_1000.json new file mode 100644 index 0000000000000000000000000000000000000000..b2929500b085b72db26cfb5a1d69a99cdbd11257 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_1000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 15569, + "1": 2431 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_1500.json b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_1500.json new file mode 100644 index 0000000000000000000000000000000000000000..6a400c73727289187267220bee7c9e522f07533c --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_1500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 15535, + "1": 2465 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..566e2cd16ed5129e2bfc35f61d1f19a06662f66b --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_2000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 15564, + "1": 2436 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_500.json b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_500.json new file mode 100644 index 0000000000000000000000000000000000000000..eb5ba877b025686efa4a661d98748a480c1ec333 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/mid_train/step_500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 382.18000000001996, + "mean_length": 3096.8, + "std_reward": 124.8400000000113, + "std_length": 1006.4000000000002, + "episode_rewards": [ + 132.49999999999739, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 1084, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 13331, + "1": 2153 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 382.18000000001996, + "mean_length": 3096.8, + "std_reward": 124.8400000000113, + "std_length": 1006.4000000000002, + "task_count": 1, + "macro_mean_reward": 382.18000000001996, + "macro_mean_length": 3096.8, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/post_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/post_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..6e50c34f3a90be8e20569b3e36b38ad3a94f4931 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/eval/post_train/step_2000.json @@ -0,0 +1,531 @@ +{ + "per_latency": { + "flappy/latency_0": { + "latency": 0, + "num_episodes": 20, + "mean_reward": 72.23999999999896, + "mean_length": 597.95, + "std_reward": 63.11439930792145, + "std_length": 509.34550896223675, + "episode_rewards": [ + 57.800000000000445, + 34.80000000000019, + 20.800000000000036, + 174.7999999999955, + 81.29999999999968, + 169.79999999999572, + 132.79999999999737, + 11.29999999999998, + 34.80000000000019, + 6.799999999999992, + 53.300000000000395, + 34.80000000000019, + 76.79999999999988, + 71.80000000000011, + 20.700000000000035, + 20.800000000000036, + 183.7999999999951, + 207.29999999999404, + 20.700000000000035, + 29.800000000000136 + ], + "episode_lengths": [ + 481, + 296, + 183, + 1426, + 671, + 1385, + 1087, + 106, + 296, + 70, + 445, + 296, + 635, + 594, + 182, + 183, + 1498, + 1688, + 182, + 255 + ], + "decoded_action_hist": { + "1": 903, + "0": 11056 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_1": { + "latency": 1, + "num_episodes": 20, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 62110, + "1": 9890 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_2": { + "latency": 2, + "num_episodes": 20, + "mean_reward": 6.615000000000006, + "mean_length": 71.75, + "std_reward": 7.366700414704039, + "std_length": 59.72342505248673, + "episode_rewards": [ + 3.8999999999999986, + 4.099999999999998, + 4.1999999999999975, + 37.20000000000022, + 4.099999999999998, + 3.8999999999999986, + 5.999999999999995, + 3.8999999999999986, + 4.299999999999997, + 3.9999999999999982, + 13.299999999999972, + 4.299999999999997, + 3.8999999999999986, + 8.999999999999984, + 3.8999999999999986, + 3.8999999999999986, + 4.099999999999998, + 4.299999999999997, + 5.999999999999995, + 3.9999999999999982 + ], + "episode_lengths": [ + 50, + 52, + 53, + 320, + 52, + 50, + 62, + 50, + 54, + 51, + 126, + 54, + 50, + 92, + 50, + 50, + 52, + 54, + 62, + 51 + ], + "decoded_action_hist": { + "0": 1186, + "1": 249 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_3": { + "latency": 3, + "num_episodes": 20, + "mean_reward": 4.689999999999996, + "mean_length": 56.55, + "std_reward": 0.6579513659838381, + "std_length": 3.9430318284284747, + "episode_rewards": [ + 3.9999999999999982, + 4.699999999999996, + 4.799999999999995, + 4.799999999999995, + 4.4999999999999964, + 3.9999999999999982, + 6.099999999999994, + 3.9999999999999982, + 4.099999999999998, + 4.399999999999997, + 4.799999999999995, + 4.699999999999996, + 4.299999999999997, + 4.799999999999995, + 6.099999999999994, + 3.9999999999999982, + 4.4999999999999964, + 4.699999999999996, + 6.099999999999994, + 4.399999999999997 + ], + "episode_lengths": [ + 51, + 58, + 59, + 59, + 56, + 51, + 63, + 51, + 52, + 55, + 59, + 58, + 54, + 59, + 63, + 51, + 56, + 58, + 63, + 55 + ], + "decoded_action_hist": { + "0": 934, + "1": 197 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_4": { + "latency": 4, + "num_episodes": 20, + "mean_reward": 5.044999999999996, + "mean_length": 60.55, + "std_reward": 1.1386285610329618, + "std_length": 8.749142815156237, + "episode_rewards": [ + 4.799999999999995, + 4.799999999999995, + 4.399999999999997, + 4.599999999999996, + 4.799999999999995, + 4.999999999999995, + 8.399999999999986, + 4.999999999999995, + 4.299999999999997, + 4.699999999999996, + 4.599999999999996, + 4.299999999999997, + 4.599999999999996, + 4.599999999999996, + 4.799999999999995, + 4.999999999999995, + 4.799999999999995, + 4.299999999999997, + 8.399999999999986, + 4.699999999999996 + ], + "episode_lengths": [ + 59, + 59, + 55, + 57, + 59, + 61, + 86, + 61, + 54, + 58, + 57, + 54, + 57, + 57, + 59, + 61, + 59, + 54, + 86, + 58 + ], + "decoded_action_hist": { + "0": 960, + "1": 251 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + } + }, + "aggregate": { + "stage": "post_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 100, + "mean_reward": 106.63800000000492, + "mean_length": 877.36, + "std_reward": 173.2974862945318, + "std_length": 1395.9831984662278, + "task_count": 1, + "macro_mean_reward": 106.6380000000049, + "macro_mean_length": 877.36, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_2_context5_2000_effbs128/eval/post_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_2_context5_2000_effbs128/eval/post_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..8536832e810f53fa683ceb4bcd805520b1688c63 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_2_context5_2000_effbs128/eval/post_train/step_2000.json @@ -0,0 +1,531 @@ +{ + "per_latency": { + "flappy/latency_0": { + "latency": 0, + "num_episodes": 20, + "mean_reward": 6.524999999999993, + "mean_length": 67.25, + "std_reward": 0.20217566619155658, + "std_length": 2.021756661915573, + "episode_rewards": [ + 5.999999999999995, + 6.499999999999993, + 6.399999999999993, + 6.5999999999999925, + 6.499999999999993, + 5.999999999999995, + 6.699999999999992, + 6.499999999999993, + 6.699999999999992, + 6.399999999999993, + 6.5999999999999925, + 6.699999999999992, + 6.699999999999992, + 6.5999999999999925, + 6.699999999999992, + 6.499999999999993, + 6.499999999999993, + 6.699999999999992, + 6.699999999999992, + 6.499999999999993 + ], + "episode_lengths": [ + 62, + 67, + 66, + 68, + 67, + 62, + 69, + 67, + 69, + 66, + 68, + 69, + 69, + 68, + 69, + 67, + 67, + 69, + 69, + 67 + ], + "decoded_action_hist": { + "0": 1265, + "1": 80 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_1": { + "latency": 1, + "num_episodes": 20, + "mean_reward": 7.48499999999999, + "mean_length": 75.95, + "std_reward": 1.9958143701256335, + "std_length": 15.367091461952063, + "episode_rewards": [ + 6.799999999999992, + 11.29999999999998, + 11.29999999999998, + 4.799999999999995, + 11.29999999999998, + 6.799999999999992, + 6.799999999999992, + 6.5999999999999925, + 6.799999999999992, + 6.799999999999992, + 4.799999999999995, + 6.799999999999992, + 6.799999999999992, + 6.799999999999992, + 6.799999999999992, + 6.799999999999992, + 11.29999999999998, + 6.799999999999992, + 6.799999999999992, + 6.699999999999992 + ], + "episode_lengths": [ + 70, + 106, + 106, + 59, + 106, + 70, + 70, + 68, + 70, + 70, + 59, + 70, + 70, + 70, + 70, + 70, + 106, + 70, + 70, + 69 + ], + "decoded_action_hist": { + "0": 1358, + "1": 161 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_2": { + "latency": 2, + "num_episodes": 20, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 57467, + "1": 14533 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_3": { + "latency": 3, + "num_episodes": 20, + "mean_reward": 4.5749999999999975, + "mean_length": 55.4, + "std_reward": 2.1463632031881215, + "std_length": 17.170905625505018, + "episode_rewards": [ + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 12.899999999999974, + 3.8999999999999986, + 8.399999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986 + ], + "episode_lengths": [ + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 122, + 50, + 86, + 50, + 50, + 50, + 50, + 50 + ], + "decoded_action_hist": { + "0": 894, + "1": 214 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_4": { + "latency": 4, + "num_episodes": 20, + "mean_reward": 3.9, + "mean_length": 50.0, + "std_reward": 1.3322676295501878e-15, + "std_length": 0.0, + "episode_rewards": [ + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986, + 3.8999999999999986 + ], + "episode_lengths": [ + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50, + 50 + ], + "decoded_action_hist": { + "0": 847, + "1": 153 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + } + }, + "aggregate": { + "stage": "post_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 100, + "mean_reward": 93.41700000000513, + "mean_length": 769.72, + "std_reward": 175.60116916183563, + "std_length": 1415.206755778109, + "task_count": 1, + "macro_mean_reward": 93.41700000000512, + "macro_mean_length": 769.72, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_1000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_1000.json new file mode 100644 index 0000000000000000000000000000000000000000..1694941fce4ad752c5bbd313ea4b99d12b5be14f --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_1000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_3": { + "latency": 3, + "num_episodes": 5, + "mean_reward": 259.74000000000865, + "mean_length": 2110.8, + "std_reward": 172.3612903177622, + "std_length": 1389.668651153936, + "episode_rewards": [ + 444.60000000002566, + 132.79999999999737, + 6.799999999999992, + 444.60000000002566, + 269.8999999999946 + ], + "episode_lengths": [ + 3600, + 1087, + 70, + 3600, + 2197 + ], + "decoded_action_hist": { + "0": 8510, + "1": 2044 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 259.74000000000865, + "mean_length": 2110.8, + "std_reward": 172.3612903177622, + "std_length": 1389.668651153936, + "task_count": 1, + "macro_mean_reward": 259.74000000000865, + "macro_mean_length": 2110.8, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_1500.json b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_1500.json new file mode 100644 index 0000000000000000000000000000000000000000..5fbf9975c67e5c0f04cf3027b14747e0f77aff6f --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_1500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_3": { + "latency": 3, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 14478, + "1": 3522 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..d844acec8c604224a6776e986a6257b56dbd06c9 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_2000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_3": { + "latency": 3, + "num_episodes": 5, + "mean_reward": 380.84000000002004, + "mean_length": 3087.0, + "std_reward": 127.52000000001118, + "std_length": 1026.0, + "episode_rewards": [ + 125.79999999999767, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 1035, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 12458, + "1": 2977 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 380.84000000002004, + "mean_length": 3087.0, + "std_reward": 127.52000000001118, + "std_length": 1026.0, + "task_count": 1, + "macro_mean_reward": 380.84000000002004, + "macro_mean_length": 3087.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_500.json b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_500.json new file mode 100644 index 0000000000000000000000000000000000000000..e00015be369ef2038cb01abbb8379a140087e2fc --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/mid_train/step_500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_3": { + "latency": 3, + "num_episodes": 5, + "mean_reward": 105.19999999999952, + "mean_length": 866.8, + "std_reward": 107.90404997033255, + "std_length": 871.505226605096, + "episode_rewards": [ + 302.4000000000004, + 92.49999999999916, + 8.399999999999986, + 115.89999999999812, + 6.799999999999992 + ], + "episode_lengths": [ + 2459, + 765, + 86, + 954, + 70 + ], + "decoded_action_hist": { + "0": 3514, + "1": 820 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 105.19999999999952, + "mean_length": 866.8, + "std_reward": 107.90404997033255, + "std_length": 871.505226605096, + "task_count": 1, + "macro_mean_reward": 105.19999999999952, + "macro_mean_length": 866.8, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/post_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/post_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..c2012b7d04e24f7a992705c4973314d293a78c75 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_3_context5_2000_effbs128/eval/post_train/step_2000.json @@ -0,0 +1,531 @@ +{ + "per_latency": { + "flappy/latency_0": { + "latency": 0, + "num_episodes": 20, + "mean_reward": 6.884999999999991, + "mean_length": 73.1, + "std_reward": 1.9938091684010226, + "std_length": 16.164467204334326, + "episode_rewards": [ + 8.899999999999984, + 3.8999999999999986, + 4.699999999999996, + 5.999999999999995, + 3.8999999999999986, + 9.099999999999984, + 6.499999999999993, + 6.199999999999994, + 4.999999999999995, + 6.5999999999999925, + 5.999999999999995, + 4.999999999999995, + 8.599999999999985, + 5.999999999999995, + 8.899999999999984, + 9.099999999999984, + 10.799999999999981, + 4.999999999999995, + 8.899999999999984, + 8.599999999999985 + ], + "episode_lengths": [ + 91, + 50, + 58, + 62, + 50, + 93, + 67, + 64, + 61, + 68, + 62, + 61, + 88, + 62, + 91, + 93, + 101, + 61, + 91, + 88 + ], + "decoded_action_hist": { + "0": 1265, + "1": 197 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_1": { + "latency": 1, + "num_episodes": 20, + "mean_reward": 8.069999999999986, + "mean_length": 82.25, + "std_reward": 3.544587423100174, + "std_length": 28.301722562416586, + "episode_rewards": [ + 9.099999999999984, + 3.8999999999999986, + 13.69999999999997, + 6.099999999999994, + 3.8999999999999986, + 13.89999999999997, + 8.999999999999984, + 14.99999999999997, + 4.799999999999995, + 9.099999999999984, + 6.099999999999994, + 4.799999999999995, + 6.799999999999992, + 6.099999999999994, + 6.799999999999992, + 14.99999999999997, + 6.5999999999999925, + 4.799999999999995, + 9.099999999999984, + 6.799999999999992 + ], + "episode_lengths": [ + 93, + 50, + 130, + 63, + 50, + 132, + 92, + 134, + 59, + 93, + 63, + 59, + 70, + 63, + 70, + 134, + 68, + 59, + 93, + 70 + ], + "decoded_action_hist": { + "0": 1391, + "1": 254 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_2": { + "latency": 2, + "num_episodes": 20, + "mean_reward": 13.190000000000015, + "mean_length": 121.75, + "std_reward": 9.237201957302931, + "std_length": 74.88850045233914, + "episode_rewards": [ + 8.399999999999986, + 25.200000000000085, + 3.8999999999999986, + 37.60000000000023, + 11.09999999999998, + 8.399999999999986, + 6.799999999999992, + 20.700000000000035, + 6.699999999999992, + 10.99999999999998, + 6.699999999999992, + 6.299999999999994, + 34.20000000000018, + 6.5999999999999925, + 15.79999999999998, + 10.599999999999982, + 6.799999999999992, + 6.299999999999994, + 15.599999999999977, + 15.09999999999997 + ], + "episode_lengths": [ + 86, + 218, + 50, + 324, + 104, + 86, + 70, + 182, + 69, + 103, + 69, + 65, + 290, + 68, + 142, + 99, + 70, + 65, + 140, + 135 + ], + "decoded_action_hist": { + "0": 2042, + "1": 393 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_3": { + "latency": 3, + "num_episodes": 20, + "mean_reward": 427.8800000000241, + "mean_length": 3465.85, + "std_reward": 69.38502432082045, + "std_length": 558.2662693554036, + "episode_rewards": [ + 125.79999999999767, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 429.000000000023, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 1035, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3482, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 55767, + "1": 13550 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_4": { + "latency": 4, + "num_episodes": 20, + "mean_reward": 7.360000000000005, + "mean_length": 78.75, + "std_reward": 7.047013551853045, + "std_length": 57.37061530086635, + "episode_rewards": [ + 3.8999999999999986, + 3.8999999999999986, + 4.799999999999995, + 4.4999999999999964, + 3.8999999999999986, + 3.8999999999999986, + 33.00000000000018, + 3.8999999999999986, + 13.99999999999997, + 3.8999999999999986, + 4.399999999999997, + 9.499999999999982, + 4.999999999999995, + 4.4999999999999964, + 4.999999999999995, + 3.8999999999999986, + 3.8999999999999986, + 8.399999999999986, + 19.000000000000025, + 3.8999999999999986 + ], + "episode_lengths": [ + 50, + 50, + 59, + 56, + 50, + 50, + 287, + 50, + 133, + 50, + 55, + 97, + 61, + 56, + 61, + 50, + 50, + 86, + 174, + 50 + ], + "decoded_action_hist": { + "0": 1160, + "1": 415 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + } + }, + "aggregate": { + "stage": "post_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 100, + "mean_reward": 92.67700000000481, + "mean_length": 764.34, + "std_reward": 170.5535950104934, + "std_length": 1374.4667709333683, + "task_count": 1, + "macro_mean_reward": 92.67700000000482, + "macro_mean_length": 764.3399999999999, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_1000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_1000.json new file mode 100644 index 0000000000000000000000000000000000000000..43a80d60d7e24e3f07367c39d16b57f1fb4caee6 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_1000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_4": { + "latency": 4, + "num_episodes": 5, + "mean_reward": 386.7400000000153, + "mean_length": 3133.8, + "std_reward": 70.92473757443024, + "std_length": 571.468424324564, + "episode_rewards": [ + 295.299999999999, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 304.60000000000065 + ], + "episode_lengths": [ + 2397, + 3600, + 3600, + 3600, + 2472 + ], + "decoded_action_hist": { + "0": 12809, + "1": 2860 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 386.7400000000153, + "mean_length": 3133.8, + "std_reward": 70.92473757443024, + "std_length": 571.468424324564, + "task_count": 1, + "macro_mean_reward": 386.7400000000153, + "macro_mean_length": 3133.8, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_1500.json b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_1500.json new file mode 100644 index 0000000000000000000000000000000000000000..24cab8acbdc696908bb8e712cdbbec2511d4672d --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_1500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_4": { + "latency": 4, + "num_episodes": 5, + "mean_reward": 416.40000000002067, + "mean_length": 3374.2, + "std_reward": 56.40000000001, + "std_length": 451.6, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 303.60000000000065 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 2471 + ], + "decoded_action_hist": { + "0": 13846, + "1": 3025 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 416.40000000002067, + "mean_length": 3374.2, + "std_reward": 56.40000000001, + "std_length": 451.6, + "task_count": 1, + "macro_mean_reward": 416.40000000002067, + "macro_mean_length": 3374.2, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..92c2af9fdcfcb604097a026c6a14e33aee00a6cf --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_2000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_4": { + "latency": 4, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 14774, + "1": 3226 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_500.json b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_500.json new file mode 100644 index 0000000000000000000000000000000000000000..0ea8cb39d78fc0ce406115bec1928b048c152c04 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/mid_train/step_500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_4": { + "latency": 4, + "num_episodes": 5, + "mean_reward": 60.739999999999384, + "mean_length": 506.8, + "std_reward": 44.601955114097024, + "std_length": 359.98744422548964, + "episode_rewards": [ + 104.29999999999866, + 50.40000000000037, + 20.200000000000028, + 120.39999999999792, + 8.399999999999986 + ], + "episode_lengths": [ + 856, + 425, + 177, + 990, + 86 + ], + "decoded_action_hist": { + "0": 2136, + "1": 398 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 60.739999999999384, + "mean_length": 506.8, + "std_reward": 44.601955114097024, + "std_length": 359.98744422548964, + "task_count": 1, + "macro_mean_reward": 60.739999999999384, + "macro_mean_length": 506.8, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/post_train/step_2000.json b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/post_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..238bf5c5e0af158d789929a507c871cdf59a73e1 --- /dev/null +++ b/evaluations/flappy/wan_oft_flappy_fix_latency_4_context5_2000_effbs128/eval/post_train/step_2000.json @@ -0,0 +1,531 @@ +{ + "per_latency": { + "flappy/latency_0": { + "latency": 0, + "num_episodes": 20, + "mean_reward": 5.569999999999995, + "mean_length": 63.55, + "std_reward": 2.3916730545791514, + "std_length": 19.043305910476782, + "episode_rewards": [ + 6.499999999999993, + 4.099999999999998, + 4.399999999999997, + 4.799999999999995, + 4.099999999999998, + 8.399999999999986, + 3.8999999999999986, + 6.5999999999999925, + 4.699999999999996, + 3.8999999999999986, + 9.499999999999982, + 4.699999999999996, + 3.8999999999999986, + 4.799999999999995, + 3.8999999999999986, + 8.399999999999986, + 4.099999999999998, + 12.899999999999974, + 3.8999999999999986, + 3.8999999999999986 + ], + "episode_lengths": [ + 67, + 52, + 55, + 59, + 52, + 86, + 50, + 68, + 58, + 50, + 97, + 58, + 50, + 59, + 50, + 86, + 52, + 122, + 50, + 50 + ], + "decoded_action_hist": { + "0": 1200, + "1": 71 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_1": { + "latency": 1, + "num_episodes": 20, + "mean_reward": 5.264999999999994, + "mean_length": 60.5, + "std_reward": 1.524557312796075, + "std_length": 10.49047186736612, + "episode_rewards": [ + 4.299999999999997, + 4.699999999999996, + 4.799999999999995, + 6.199999999999994, + 4.699999999999996, + 4.4999999999999964, + 4.599999999999996, + 4.1999999999999975, + 6.499999999999993, + 4.399999999999997, + 6.199999999999994, + 6.499999999999993, + 4.1999999999999975, + 6.199999999999994, + 4.1999999999999975, + 4.4999999999999964, + 4.699999999999996, + 10.89999999999998, + 4.599999999999996, + 4.399999999999997 + ], + "episode_lengths": [ + 54, + 58, + 59, + 64, + 58, + 56, + 57, + 53, + 67, + 55, + 64, + 67, + 53, + 64, + 53, + 56, + 58, + 102, + 57, + 55 + ], + "decoded_action_hist": { + "0": 1083, + "1": 127 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_2": { + "latency": 2, + "num_episodes": 20, + "mean_reward": 5.4649999999999945, + "mean_length": 62.05, + "std_reward": 0.8956980518009389, + "std_length": 4.831925082200675, + "episode_rewards": [ + 4.699999999999996, + 5.999999999999995, + 4.999999999999995, + 6.5999999999999925, + 4.899999999999995, + 4.899999999999995, + 4.4999999999999964, + 4.899999999999995, + 6.799999999999992, + 4.799999999999995, + 6.5999999999999925, + 6.799999999999992, + 4.699999999999996, + 6.5999999999999925, + 4.699999999999996, + 4.699999999999996, + 5.999999999999995, + 6.799999999999992, + 4.4999999999999964, + 4.799999999999995 + ], + "episode_lengths": [ + 58, + 62, + 61, + 68, + 60, + 60, + 56, + 60, + 70, + 59, + 68, + 70, + 58, + 68, + 58, + 58, + 62, + 70, + 56, + 59 + ], + "decoded_action_hist": { + "0": 1082, + "1": 159 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_3": { + "latency": 3, + "num_episodes": 20, + "mean_reward": 7.604999999999999, + "mean_length": 79.4, + "std_reward": 5.2202945319206115, + "std_length": 41.852598485637664, + "episode_rewards": [ + 4.699999999999996, + 6.799999999999992, + 6.5999999999999925, + 6.799999999999992, + 6.799999999999992, + 4.699999999999996, + 4.799999999999995, + 4.699999999999996, + 27.80000000000012, + 4.999999999999995, + 6.799999999999992, + 9.199999999999983, + 4.899999999999995, + 6.799999999999992, + 4.899999999999995, + 4.699999999999996, + 6.799999999999992, + 9.099999999999984, + 15.19999999999997, + 4.999999999999995 + ], + "episode_lengths": [ + 58, + 70, + 68, + 70, + 70, + 58, + 59, + 58, + 244, + 61, + 70, + 94, + 60, + 70, + 60, + 58, + 70, + 93, + 136, + 61 + ], + "decoded_action_hist": { + "0": 1361, + "1": 227 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_4": { + "latency": 4, + "num_episodes": 20, + "mean_reward": 413.42000000002145, + "mean_length": 3349.25, + "std_reward": 77.46433760125731, + "std_length": 623.3434747392484, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 202.39999999999426, + 317.4000000000031, + 190.39999999999478, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 1648, + 2582, + 1555, + 3600 + ], + "decoded_action_hist": { + "0": 54994, + "1": 11991 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + } + }, + "aggregate": { + "stage": "post_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 100, + "mean_reward": 87.46500000000428, + "mean_length": 722.95, + "std_reward": 166.64257221671508, + "std_length": 1342.5982450085357, + "task_count": 1, + "macro_mean_reward": 87.46500000000428, + "macro_mean_length": 722.95, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/copy_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/copy_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..3916d8d1f7cb770ccbddf2c9057f5041762a0892 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/copy_verification.json @@ -0,0 +1,5 @@ +{ + "source_files_verified": 735, + "preexisting_files_preserved": 1435, + "identities_match": true +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/data_readability_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/data_readability_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..7eba7621cfc5b9db373a56ade0feb14e63d7b3d9 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/data_readability_verification.json @@ -0,0 +1,7050 @@ +[ + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_2", + "source_revision": "df04b6f30856b2903e65a6a870f7440055f02311", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_2/data/train-00000-of-00001.parquet", + "identity": "d7f14fe96cc13b1cad520b3e9d1c51ae675773a66183ada6bae5bca965e7169f", + "size": 38515375, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_2/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2311, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 29, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "latency": 2, + "latency_ms": 57.142857142857146, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 200000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "f90d1dac8b31a18c0496884729c744d7a4c877ae", + "source_path": "val.parquet", + "target_repo": "latency-sensitive-bench/memory-data", + "type": "dataset", + "path": "training/flappy_fixed_latency_3_200ep_7k2steps_ghost15/val.parquet", + "identity": "d8017b5d3c4480372e466f0860b69e34031b90edbc44148b95f6f28ad0aae560", + "size": 264981651, + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/memory-data", + "path": "training/flappy_fixed_latency_3_200ep_7k2steps_ghost15/val.parquet", + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd" + } + ], + "parquet_rows": 14400, + "row_groups": 8, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 127, + "latency_ms": 100.0, + "latency_raw_frames": 3, + "raw_reward": 0.10000000149011612, + "seed": 127, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fixed_l2_fs4", + "source_revision": "1bdd2ea6ea7db07139e4694f445c12198399c2b8", + "source_path": "train.parquet", + "target_repo": "latency-sensitive-bench/vla-finetuning-scope-data", + "type": "dataset", + "path": "training/flappy_fixed_l2_fs4/train.parquet", + "identity": "00a0b4f862e5a1e6afda14bcb6106c7a4ebe19b09ef8f131c02554bb2e63de84", + "size": 11126190209, + "revision": "1354ac73e35d37849be824ccf2dc106796f0c3d0", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/vla-finetuning-scope-data", + "path": "training/flappy_fixed_l2_fs4/train.parquet", + "revision": "1354ac73e35d37849be824ccf2dc106796f0c3d0" + } + ], + "parquet_rows": 948645, + "row_groups": 3706, + "columns": [ + "action", + "action_id", + "action_text", + "deterministic", + "done", + "episode_idx", + "image_stack", + "latency_ms", + "latency_raw_frames", + "prompt", + "reward", + "seed", + "split", + "t" + ], + "decoded_images": 4, + "first_row_scalars": { + "action_id": 0, + "action_text": "noop", + "deterministic": true, + "done": false, + "episode_idx": 0, + "latency_ms": 66.66666666666667, + "latency_raw_frames": 2, + "reward": 0.20000000298023224, + "seed": 1, + "split": "train", + "t": 0 + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_7", + "source_revision": "d04b9bc42bfbb7da400438a9fb75c49b198f7776", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_7/data/val-00000-of-00001.parquet", + "identity": "7c81db68735e8b35351e7c5882a2ee3c23d92a703c280ba2ed92cdbfaa8a2f6e", + "size": 3750129, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_7/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 225, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 7, + "latency_ms": 200.0, + "reward": 0.0, + "done": false, + "episode_idx": 15, + "t": 0, + "seed": 700015, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "identity": "605cd50b8ad5f520756a9190eb7c30dcd58c4d35330ac388e1823f8d09158b32", + "size": 304119714, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + }, + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 49147, + "row_groups": 4, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 0, + "decision_step": 0, + "split": "train", + "seed": 25, + "latency_raw_frames": 2, + "latency_ms": 80.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_4_1000ep_7k2steps/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_4_1000ep_7k2steps/train-00000-of-00001.parquet", + "identity": "2f1b638c23c05324939ee81e8d14f4351ef5b7ce8dd59cebac5c42d4f9fc9c1c", + "size": 207641775, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_4_1000ep_7k2steps/train-00000-of-00001.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + }, + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_4_1000ep_7k2steps/train-00000-of-00001.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 30304, + "row_groups": 3, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 0, + "decision_step": 0, + "split": "train", + "seed": 12, + "latency_raw_frames": 4, + "latency_ms": 160.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_0_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_0_200ep/val.parquet", + "identity": "900cebe2cd7ac9be11c1c80f1f7f0abf7243bb610d555567f006069b8b4a83c6", + "size": 39825924, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_0_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_0_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 68014, + "row_groups": 266, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 5, + "action_text": "LEFTFIRE", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 3, + "latency_ms": 0.0, + "latency_raw_frames": 0, + "raw_reward": 0.0, + "seed": 3, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_4_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_4_200ep/train.parquet", + "identity": "cbc9ea8ecc830e2f15042ef12ac52577c8db04d593047aec1e30e7f110f933a1", + "size": 194112105, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_4_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_4_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 332733, + "row_groups": 1300, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 3, + "action_text": "LEFT", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 0, + "latency_ms": 266.6666666666667, + "latency_raw_frames": 16, + "raw_reward": 0.0, + "seed": 5, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_6_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_6_200ep/val.parquet", + "identity": "1becbfeca092153c722db7d810ae5663e916761e4d143d22494acf3ad000cc5f", + "size": 30637048, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_6_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_6_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 54672, + "row_groups": 214, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 1, + "action_text": "FIRE", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 3, + "latency_ms": 400.0, + "latency_raw_frames": 24, + "raw_reward": 0.0, + "seed": 1, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_3", + "source_revision": "709111a3b59e413214e0417bd878291734ec9427", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_3/data/train-00000-of-00001.parquet", + "identity": "75a1111b798cd06fe0d96930630925760d199b6094a73b68f0a91ff85772eb1b", + "size": 201900030, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_3/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 283754, + "row_groups": 3, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 1, + "action_text": "FIRE", + "latency": 3, + "latency_ms": 200.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 300000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_1", + "source_revision": "e09b7f2c51726bb169d827fe973b2d429b6b1432", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_1/data/val-00000-of-00001.parquet", + "identity": "ab8353529f419e970833e4e87ca3c0ccad5dd3e679e461e3472eccadc943b729", + "size": 3670413, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_1/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 218, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 29, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "latency": 1, + "latency_ms": 28.571428571428573, + "reward": 0.0, + "done": false, + "episode_idx": 17, + "t": 0, + "seed": 100017, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_2_small", + "source_revision": "c9bbd1744020a587b55590c897628510e0741136", + "source_path": "data/train-00000-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_2_small/data/train-00000-of-00003.parquet", + "identity": "e3bc4f36d1b3776460f5cfc56254bd780a51ca77f0a5768fa71067235877dbf6", + "size": 378685939, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_2_small/data/train-00000-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 2, + "latency_ms": 66.66666666666667, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 2, + "t": 0, + "seed": 200002, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_2_small", + "source_revision": "c9bbd1744020a587b55590c897628510e0741136", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_2_small/data/val-00000-of-00001.parquet", + "identity": "444e4947d656feeae9da17c65f70d41764bf3b9800b4f2e5e8cc6d63d0f9205c", + "size": 127742697, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_2_small/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 10800, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 2, + "latency_ms": 66.66666666666667, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 200000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00000-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00000-of-00021.parquet", + "identity": "cb43d1d8398cc4f29490c5515c467250314055b7326d9ef0c72fca23fbe93a1b", + "size": 454398953, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00000-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 0, + "decision_step": 0, + "split": "train", + "seed": 10, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 0, + "action_text": "push_left_hard" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00003-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00003-of-00021.parquet", + "identity": "6ba4e2cd887243e5b95c871c147f20793e93c3d3437b2ffa93b1ef1b2761af4d", + "size": 454310544, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00003-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 28, + "decision_step": 2058, + "split": "train", + "seed": 38, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 3, + "action_text": "push_right_soft" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00006-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00006-of-00021.parquet", + "identity": "6358ed328905661a71fe0e6b3b7e7e776c671b32f0cb6e988276ab267a417c27", + "size": 455593560, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00006-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 56, + "decision_step": 4116, + "split": "train", + "seed": 66, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00009-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00009-of-00021.parquet", + "identity": "e9b2dd318410792152b5dc0fd8b0a4166bebdc92bb6b56adc3868fc44161c9c2", + "size": 456214246, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00009-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 84, + "decision_step": 6174, + "split": "train", + "seed": 94, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 3, + "action_text": "push_right_soft" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00012-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00012-of-00021.parquet", + "identity": "0d6ad9e21e611f923b2d6bccd6ad0d3753439ed9faf15debe0c9a66f4c00660b", + "size": 457128169, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00012-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 113, + "decision_step": 1032, + "split": "train", + "seed": 123, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 1, + "action_text": "push_left_soft" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00015-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00015-of-00021.parquet", + "identity": "e833e89c4d4a787c8a6903bbab07b7339ebc35f40fb4fd38ddaca97c0ccf5549", + "size": 456465862, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00015-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67885, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 142, + "decision_step": 3090, + "split": "train", + "seed": 152, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00018-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00018-of-00021.parquet", + "identity": "2396f259cf312ba74ae36eb6f9031de2a8e53764a364d2cc6e3133ea187b7f51", + "size": 457221367, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00018-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67885, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 171, + "decision_step": 5145, + "split": "train", + "seed": 181, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/val-00000-of-00001.parquet", + "identity": "92e4ebaee8c7e73bb0b960eee6577bb22e3db35b26b77a47344a843159ac8804", + "size": 96309898, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/val-00000-of-00001.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 14400, + "row_groups": 2, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 127, + "decision_step": 0, + "split": "val", + "seed": 137, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 0, + "action_text": "push_left_hard" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_3_small", + "source_revision": "8470ae1f00ca10bb8646ac7cfdccf9682e053eab", + "source_path": "data/train-00002-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_3_small/data/train-00002-of-00003.parquet", + "identity": "d318f64efc3cfa028d5a09a635acfe9c49572d7b1d4649dea5e990776a883cc5", + "size": 381935652, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_3_small/data/train-00002-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 3, + "latency_ms": 100.0, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 2400, + "seed": 300000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_13", + "source_revision": "2a5c5d2aa4730db22fd33f4f00f4d93fc5043465", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_13/data/val-00000-of-00001.parquet", + "identity": "ee41ccec8dc6c18bc43c7cab6a340d6b2a6fc6b79be271822a49cfe16864162f", + "size": 27776071, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_13/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 40524, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 13, + "latency_ms": 866.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 9, + "t": 0, + "seed": 1300009, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_14", + "source_revision": "9867d03ad02a4d025cb6cc88281f95596f83a814", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_14/data/train-00000-of-00001.parquet", + "identity": "74ba8955e7262b1bfd3601624e0a7501c34df8f5030f62f093d15393d664d21c", + "size": 103891596, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_14/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 150406, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 14, + "latency_ms": 933.3333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1400000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_15", + "source_revision": "21811c6039f0a538f11821730ee6792bd0d6c0b4", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_15/data/val-00000-of-00001.parquet", + "identity": "bf90843b70be4d24f9a2946aa803357741ffe80000e2a5eb43ef4edf721657ff", + "size": 21496501, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_15/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 31189, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 15, + "latency_ms": 1000.0000000000001, + "reward": 0.0, + "done": false, + "episode_idx": 8, + "t": 0, + "seed": 1500008, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_12", + "source_revision": "258c4ae285ad34445578dea4e786c408fcd6be30", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_12/data/train-00000-of-00001.parquet", + "identity": "c512c87ab13f1fcde4dd1cd013361d0e2f886b1c895df5bb9e163bc47a821505", + "size": 40638421, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_12/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2501, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 34, + "action_text": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "latency": 12, + "latency_ms": 342.85714285714283, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1200000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_0_small", + "source_revision": "7e974f1e8e1ad7ca4ee0ffad0849d886d64128d3", + "source_path": "data/train-00001-of-00002.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_0_small/data/train-00001-of-00002.parquet", + "identity": "62ccc7011c5fbdc36147142ab009f453f229c3fe8a9a968885eb69fe5efc0fc6", + "size": 552434878, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_0_small/data/train-00001-of-00002.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 46943, + "row_groups": 5, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 1, + "action_text": "FLAP", + "latency": 0, + "latency_ms": 0.0, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 15, + "t": 1794, + "seed": 15, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_10", + "source_revision": "94762ebd79edb55d6e83ba2046729e3e6574d64d", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_10/data/val-00000-of-00001.parquet", + "identity": "25854a5f15dfcdcd3405bc549f238bbe0b7af55aa15f041b82870c911a19be5c", + "size": 4368715, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_10/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 271, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 10, + "latency_ms": 285.7142857142857, + "reward": 0.0, + "done": false, + "episode_idx": 3, + "t": 0, + "seed": 1000003, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "source_revision": "862b6cad9b2c87dc129d5cf45a9b23b18450c4ef", + "source_path": "train.parquet", + "target_repo": "latency-sensitive-bench/memory-data", + "type": "dataset", + "path": "training/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15/train.parquet", + "identity": "ef3d65a30d0e9222cf7e688de509821fd7697585bacd56ed7c54084773b3d756", + "size": 986794901, + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/memory-data", + "path": "training/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15/train.parquet", + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd" + } + ], + "parquet_rows": 559520, + "row_groups": 2186, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 3, + "action_text": "LEFT", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 0, + "latency_ms": 400.0, + "latency_raw_frames": 24, + "raw_reward": 0.0, + "seed": 6, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_8", + "source_revision": "e02c6b1db39b37965569d4f559f1d710fb55c58a", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_8/data/val-00000-of-00001.parquet", + "identity": "1308823b4eb6585dd75c93a2898ad0a268e8d590a39c33618676dd5e3b15edf4", + "size": 31403424, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_8/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 45096, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 1, + "action_text": "FIRE", + "latency": 8, + "latency_ms": 533.3333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 1, + "t": 0, + "seed": 800001, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_1_small", + "source_revision": "749a849911b8c35005c26241fb4edbcee1286832", + "source_path": "data/train-00000-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_1_small/data/train-00000-of-00003.parquet", + "identity": "88186ce5dae7b00ffaf5518741f9f09190cc991c8bbc483a6567c5622575602f", + "size": 377996230, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_1_small/data/train-00000-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 1, + "action_text": "FLAP", + "latency": 1, + "latency_ms": 33.333333333333336, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 100000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_1_small", + "source_revision": "749a849911b8c35005c26241fb4edbcee1286832", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_1_small/data/val-00000-of-00001.parquet", + "identity": "646c16683b87381e239f9072418e6598c4e15bb7bf155c0fbc74163e899e274c", + "size": 126807978, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_1_small/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 10800, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 1, + "action_text": "FLAP", + "latency": 1, + "latency_ms": 33.333333333333336, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 10, + "t": 0, + "seed": 100010, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_4_small", + "source_revision": "8a6a091570530476b313976b6a8d1b94d600f6d9", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_4_small/data/val-00000-of-00001.parquet", + "identity": "bf90759628eeede82827cc6db03b8963c7de7ab39ef2acac436aa1a3dcdb0299", + "size": 128069769, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_4_small/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 10800, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 4, + "latency_ms": 133.33333333333334, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 11, + "t": 0, + "seed": 400011, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_7", + "source_revision": "d76a57ef1b84c591c47975347c539da59670f3b4", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_7/data/train-00000-of-00001.parquet", + "identity": "8b453a5e659d7bbefea1a7c769b2b5e9e26fc0632dc57ab81434128e8039ab07", + "size": 164751658, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_7/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 233995, + "row_groups": 3, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 7, + "latency_ms": 466.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 700000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_10", + "source_revision": "9a3f7f736402b063885068e683860b6746e5337d", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_10/data/val-00000-of-00001.parquet", + "identity": "a63c50ab4fba0b07cd1c9c747979f77436740a57688efda8a3da14fec76c7564", + "size": 27397693, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_10/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 39148, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 10, + "latency_ms": 666.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 1, + "t": 0, + "seed": 1000001, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_8", + "source_revision": "164f4e56abb6c2637801e779918a99a7d70c9c3a", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_8/data/train-00000-of-00001.parquet", + "identity": "180b4a69f7a51dac61be4f3387d976919f7db2ede01e455df59106ca1df1cd16", + "size": 27577965, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_8/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 1678, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 8, + "latency_ms": 228.57142857142858, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 800000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_0_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_0_200ep/val.parquet", + "identity": "1ee8de7f4c4c40760fb02789fb7739de6be58fd16791082ebcc8e4092ac870c5", + "size": 818931777, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_0_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_0_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 72000, + "row_groups": 282, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 3, + "latency_ms": 0.0, + "latency_raw_frames": 0, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 0 raw frames (0.00 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 3, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_2_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_2_200ep/train.parquet", + "identity": "5b3cc78f212f2b1c4569aa73f091e43c2a0deec2297b78988dea7df9ec5e2d44", + "size": 3786489065, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_2_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_2_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 330734, + "row_groups": 1292, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 2, + "env_name": "flappy", + "episode_idx": 0, + "latency_ms": 66.66666666666667, + "latency_raw_frames": 2, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 2 raw frames (66.67 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 0, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_3_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_3_200ep/val.parquet", + "identity": "bf13854730d744f2e38ff246046191d472dc2e1b28923ded15949490b9cceb15", + "size": 820291587, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_3_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_3_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 72000, + "row_groups": 282, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 3, + "latency_ms": 100.0, + "latency_raw_frames": 3, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 3 raw frames (100.00 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 521, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_4", + "source_revision": "4dee6c0fd4abb86507001c4dc7404a5310c2f30a", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_4/data/train-00000-of-00001.parquet", + "identity": "d8dde9ed68760c13a1cb42261477cf36c66cf950347f913f44b7034cff7a8670", + "size": 38939107, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_4/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2345, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 4, + "latency_ms": 114.28571428571429, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 400000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_12", + "source_revision": "ac9d21a59b908e717de2bee2cc749b90495a00b8", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_12/data/val-00000-of-00001.parquet", + "identity": "025714facc9c5f90b2c052ab37301212dac788238715dccefccb4f5198564062", + "size": 30575706, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_12/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 44527, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 12, + "latency_ms": 800.0, + "reward": 0.0, + "done": false, + "episode_idx": 7, + "t": 0, + "seed": 1200007, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_0", + "source_revision": "2deacb435543c52ae51f3c947bf917fdb0b5493c", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_0/data/train-00000-of-00001.parquet", + "identity": "2833b07e87f51bd81ae41702af0247bb188f9eda3afc8a296b8e718f6d5f0585", + "size": 229012139, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_0/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 324000, + "row_groups": 3, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 4, + "action_text": "RIGHTFIRE", + "latency": 0, + "latency_ms": 0.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 0, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_5", + "source_revision": "f614a91258d86cb34cd2e3e0e79e2bf69b321e76", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_5/data/val-00000-of-00001.parquet", + "identity": "0ed632b621ce5895fb4d0e1436919543f3e3ec2bea59ebc0c5cb35c729db3399", + "size": 4348336, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_5/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 260, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 5, + "latency_ms": 142.85714285714286, + "reward": 0.0, + "done": false, + "episode_idx": 10, + "t": 0, + "seed": 500010, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_14", + "source_revision": "fb78a1bc6280a4d0a6a60e8231d63d848819bd1e", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_14/data/train-00000-of-00001.parquet", + "identity": "4d8bc75ef45e8a4f6755450c43cf408d3c0d857e5fed06abf9642cac43ec1c90", + "size": 36779288, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_14/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2325, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 28, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "latency": 14, + "latency_ms": 400.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1400000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_6", + "source_revision": "0766ba4ea9196b861302e99e564b7d2e05453885", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_6/data/val-00000-of-00001.parquet", + "identity": "7b7c74e26f404e06c4f260654c881bad82d8487c5fa8818f8dc43f3aee999619", + "size": 41350548, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_6/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 58711, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 3, + "action_text": "LEFT", + "latency": 6, + "latency_ms": 400.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 600000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_2", + "source_revision": "df04b6f30856b2903e65a6a870f7440055f02311", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_2/data/val-00000-of-00001.parquet", + "identity": "f2e3e109491679120505c1703270fd5f05e7213c3e69db561260f4ba0c9911d4", + "size": 4147651, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_2/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 246, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 29, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "latency": 2, + "latency_ms": 57.142857142857146, + "reward": 0.0, + "done": false, + "episode_idx": 2, + "t": 0, + "seed": 200002, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_11", + "source_revision": "78e4e1b6d9bb57c5f30712cd0ed2f3b02877ff56", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_11/data/train-00000-of-00001.parquet", + "identity": "8d144ba5a52d82b0ea4386c59b4dbffa8c32ce96090e3df8f19a4abe63f7b5e0", + "size": 74320602, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_11/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 107332, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 11, + "latency_ms": 733.3333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1100000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fixed_l2_fs4", + "source_revision": "1bdd2ea6ea7db07139e4694f445c12198399c2b8", + "source_path": "val.parquet", + "target_repo": "latency-sensitive-bench/vla-finetuning-scope-data", + "type": "dataset", + "path": "training/flappy_fixed_l2_fs4/val.parquet", + "identity": "97ea2116d795c4a109d0163ff96df7f1e1b8d4fd5232469a93c4190b4da3ebd7", + "size": 117484928, + "revision": "1354ac73e35d37849be824ccf2dc106796f0c3d0", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/vla-finetuning-scope-data", + "path": "training/flappy_fixed_l2_fs4/val.parquet", + "revision": "1354ac73e35d37849be824ccf2dc106796f0c3d0" + } + ], + "parquet_rows": 10000, + "row_groups": 40, + "columns": [ + "action", + "action_id", + "action_text", + "deterministic", + "done", + "episode_idx", + "image_stack", + "latency_ms", + "latency_raw_frames", + "prompt", + "reward", + "seed", + "split", + "t" + ], + "decoded_images": 4, + "first_row_scalars": { + "action_id": 0, + "action_text": "noop", + "deterministic": true, + "done": false, + "episode_idx": 16, + "latency_ms": 66.66666666666667, + "latency_raw_frames": 2, + "reward": 0.20000000298023224, + "seed": 17, + "split": "val", + "t": 0 + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_0_100ep_1ksteps_og/train.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_0_100ep_1ksteps_og/train.parquet", + "identity": "729e045c8f5ece49f1b776f342acf733a4dc62883aa19f203690d3cfe75782a4", + "size": 280452562, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_0_100ep_1ksteps_og/train.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 90000, + "row_groups": 352, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 2, + "action_text": "zero_force", + "decision_step": 0, + "env_name": "inverted_pendulum_rgb_state", + "episode_idx": 0, + "latency_ms": 0.0, + "latency_raw_frames": 0, + "raw_reward": 1.0, + "seed": 0, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00001-of-00002.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00001-of-00002.parquet", + "identity": "7bef0055f103c530c8a3ae2713650533573f1fe971f10ec44d9051a3ffdeee01", + "size": 303262421, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00001-of-00002.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + }, + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00001-of-00002.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 49146, + "row_groups": 4, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 503, + "decision_step": 9, + "split": "train", + "seed": 1636, + "latency_raw_frames": 2, + "latency_ms": 80.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_4_1000ep_7k2steps/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_4_1000ep_7k2steps/val-00000-of-00001.parquet", + "identity": "815982b8d4aa76a8db672c1e2165a29cae39562ad82f1562fecb8c0dd375f0f0", + "size": 2012800, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_4_1000ep_7k2steps/val-00000-of-00001.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + }, + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_4_1000ep_7k2steps/val-00000-of-00001.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 292, + "row_groups": 1, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 16, + "decision_step": 0, + "split": "val", + "seed": 32, + "latency_raw_frames": 4, + "latency_ms": 160.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_2_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_2_200ep/train.parquet", + "identity": "ed5ffa693f9dbbabc448992bd5b9cafcfc1c94fa7e173b0b3653e27d3699c552", + "size": 204947383, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_2_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_2_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 349558, + "row_groups": 1366, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 2, + "action_text": "RIGHT", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 0, + "latency_ms": 133.33333333333334, + "latency_raw_frames": 8, + "raw_reward": 0.0, + "seed": 1, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_4_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_4_200ep/val.parquet", + "identity": "8ea8cd47267494ef55f4d5e10595b5b4b2e73a51992f2fdcc910687c50fb6507", + "size": 35026048, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_4_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_4_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 60809, + "row_groups": 238, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 3, + "action_text": "LEFT", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 3, + "latency_ms": 266.6666666666667, + "latency_raw_frames": 16, + "raw_reward": 0.0, + "seed": 54, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_8_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_8_200ep/train.parquet", + "identity": "04efa8cc7c30b52b8820a057825194edc793fbb20e0e2c214f09d11eb175fb2c", + "size": 159016885, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_8_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_8_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 282142, + "row_groups": 1103, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 1, + "action_text": "FIRE", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 0, + "latency_ms": 533.3333333333334, + "latency_raw_frames": 32, + "raw_reward": 0.0, + "seed": 23, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_3", + "source_revision": "709111a3b59e413214e0417bd878291734ec9427", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_3/data/val-00000-of-00001.parquet", + "identity": "2ed7e485d9b8b39810880155bbbf498683c969401c03a7a91e5fd89feae71ffd", + "size": 49869161, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_3/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 70055, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 1, + "action_text": "FIRE", + "latency": 3, + "latency_ms": 200.0, + "reward": 0.0, + "done": false, + "episode_idx": 6, + "t": 0, + "seed": 300006, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_9", + "source_revision": "3405ac7c12ca266e2d3e83047bbd06e1e5eed220", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_9/data/train-00000-of-00001.parquet", + "identity": "6a11c4f113f2aa1607d1c0adae18c7a548cbb66c71722aa7edb8ba5596acaea0", + "size": 28986478, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_9/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 1755, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 9, + "latency_ms": 257.14285714285717, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 900000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_2_small", + "source_revision": "c9bbd1744020a587b55590c897628510e0741136", + "source_path": "data/train-00001-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_2_small/data/train-00001-of-00003.parquet", + "identity": "5c6101a32b97e11bb8c5b89bad21b11bb6c88399b16de7d65bf817d8d01fa2c7", + "size": 383763664, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_2_small/data/train-00001-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 2, + "latency_ms": 66.66666666666667, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 2, + "t": 1200, + "seed": 200002, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_13", + "source_revision": "dbb7678df6396735290099d070dfb5827a4cf4a4", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_13/data/train-00000-of-00001.parquet", + "identity": "d3a51e43947c9d2c66437fa8a4acbcf3c05abc3dc4c1f87b7f63ab8c703bed96", + "size": 39227814, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_13/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2442, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 13, + "latency_ms": 371.42857142857144, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1300000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00001-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00001-of-00021.parquet", + "identity": "8afe63c329537f8fa62eb0d104a68e7b4e4208a57d26e2d6ebce3362f6071789", + "size": 455686132, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00001-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 9, + "decision_step": 3086, + "split": "train", + "seed": 19, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00004-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00004-of-00021.parquet", + "identity": "8e443276e0d214e5ec80e7a5d2efc6c1d5b99da9a1c7ab698449b342e552068b", + "size": 455557331, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00004-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 37, + "decision_step": 5144, + "split": "train", + "seed": 47, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00007-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00007-of-00021.parquet", + "identity": "a3643331e2a0f7f3d0b5b9c3366e886815e0a9de34325beb0ff5842f4cdabf80", + "size": 454756348, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00007-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 66, + "decision_step": 2, + "split": "train", + "seed": 76, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00010-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00010-of-00021.parquet", + "identity": "9757fe738e5fa7d9c440387e6b2c0cc9e4d1527f3c5fef53934a08823366c99a", + "size": 453774379, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00010-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 94, + "decision_step": 2060, + "split": "train", + "seed": 104, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00013-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00013-of-00021.parquet", + "identity": "56a8f124b7fda076381f4378351de3601578597bfe925b6923d533fb0aee4a61", + "size": 454824005, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00013-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 122, + "decision_step": 4118, + "split": "train", + "seed": 132, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00016-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00016-of-00021.parquet", + "identity": "6ae1949b5c3a9054b46f453fd41fa23c0c598d2543dd3fd282a3764aa70a3afc", + "size": 455725654, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00016-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67885, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 151, + "decision_step": 6175, + "split": "train", + "seed": 161, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 3, + "action_text": "push_right_soft" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00019-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00019-of-00021.parquet", + "identity": "20d8814531c465f6f5169ded806690c8cd71d16dbc753225cd8cf991b3b1aa17", + "size": 456448614, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00019-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67885, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 181, + "decision_step": 1030, + "split": "train", + "seed": 191, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_3_small", + "source_revision": "8470ae1f00ca10bb8646ac7cfdccf9682e053eab", + "source_path": "data/train-00000-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_3_small/data/train-00000-of-00003.parquet", + "identity": "dadeed8e7ed2d008b42f4c6314c366a754e0059626b0c3f72ce2a4687b562987", + "size": 380525032, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_3_small/data/train-00000-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 3, + "latency_ms": 100.0, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 300000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_3_small", + "source_revision": "8470ae1f00ca10bb8646ac7cfdccf9682e053eab", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_3_small/data/val-00000-of-00001.parquet", + "identity": "8b5b93a55d77bbdecbb2f0215792dd6cbb8dd0ddb2066e71ccd0b25b8f57fc43", + "size": 126830866, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_3_small/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 10800, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 3, + "latency_ms": 100.0, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 5, + "t": 0, + "seed": 300005, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_2", + "source_revision": "c0605fa1df28b19249849c3e748bbc445bef6526", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_2/data/train-00000-of-00001.parquet", + "identity": "9b722b9c21720771e75b7c5ec03cf22844b7dc46dfc6199ffe295028d6846551", + "size": 205972739, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_2/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 288000, + "row_groups": 3, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 2, + "action_text": "RIGHT", + "latency": 2, + "latency_ms": 133.33333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 1, + "t": 0, + "seed": 200001, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_14", + "source_revision": "9867d03ad02a4d025cb6cc88281f95596f83a814", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_14/data/val-00000-of-00001.parquet", + "identity": "00501c7df47441edd52cce449637bacbb3abb5f9f2342e9b4bd1ef405c467bc2", + "size": 25042544, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_14/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 36184, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 14, + "latency_ms": 933.3333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 2, + "t": 0, + "seed": 1400002, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_9", + "source_revision": "7be93d38340141e385913e540b028024d74f1cf8", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_9/data/train-00000-of-00001.parquet", + "identity": "488f3c85944f9774713820447f4059977a58bc0ddf064f397c9fe22bbb55d46f", + "size": 112694760, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_9/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 165527, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 9, + "latency_ms": 600.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 900000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_12", + "source_revision": "258c4ae285ad34445578dea4e786c408fcd6be30", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_12/data/val-00000-of-00001.parquet", + "identity": "c7e49bd90866da520bf9120a16ec9aa197cd493e5c3ba79379dbbb44dfa72c8e", + "size": 4148005, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_12/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 253, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 34, + "action_text": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "latency": 12, + "latency_ms": 342.85714285714283, + "reward": 0.0, + "done": false, + "episode_idx": 14, + "t": 0, + "seed": 1200014, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_0_small", + "source_revision": "7e974f1e8e1ad7ca4ee0ffad0849d886d64128d3", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_0_small/data/val-00000-of-00001.parquet", + "identity": "c33fcddbc4ddf7a46f89916ef3b82a16a60558d4270ff70166bb4c9e11eb88d8", + "size": 127044995, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_0_small/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 10800, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 0, + "latency_ms": 0.0, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 2, + "t": 0, + "seed": 2, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_1", + "source_revision": "7190bf45dfface92ec7cbaa3c774a7e29616db52", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_1/data/train-00000-of-00001.parquet", + "identity": "52490692cb2e2ad82f9e2673ab7bc5e58b67a11907c3520d26b367f8dd34b42d", + "size": 205304187, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_1/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 288000, + "row_groups": 3, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 2, + "action_text": "RIGHT", + "latency": 1, + "latency_ms": 66.66666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 100000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "source_revision": "862b6cad9b2c87dc129d5cf45a9b23b18450c4ef", + "source_path": "val.parquet", + "target_repo": "latency-sensitive-bench/memory-data", + "type": "dataset", + "path": "training/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15/val.parquet", + "identity": "1fde0d1e2a9d3adf2447cd9ac35c8d9e63d4c4b867d1bcc65a31309c04847735", + "size": 6813323, + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/memory-data", + "path": "training/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15/val.parquet", + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd" + } + ], + "parquet_rows": 3943, + "row_groups": 16, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 3, + "action_text": "LEFT", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 127, + "latency_ms": 400.0, + "latency_raw_frames": 24, + "raw_reward": 0.0, + "seed": 127, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_15", + "source_revision": "c586ddd84f76b6295de7929b62ad4da3359b0348", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_15/data/train-00000-of-00001.parquet", + "identity": "a8a85d51855c572c6906523a6e3a09e671b5db3da9660b6a9214e8e7b33b677c", + "size": 32298813, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_15/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2061, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 15, + "latency_ms": 428.57142857142856, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1500000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_1_small", + "source_revision": "749a849911b8c35005c26241fb4edbcee1286832", + "source_path": "data/train-00001-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_1_small/data/train-00001-of-00003.parquet", + "identity": "f5e611bb50bf679ee98706b97d4f9b921476f2becd9ffc7019e881d99e4c9a1c", + "size": 382449970, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_1_small/data/train-00001-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 1, + "latency_ms": 33.333333333333336, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 1200, + "seed": 100000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_4_small", + "source_revision": "8a6a091570530476b313976b6a8d1b94d600f6d9", + "source_path": "data/train-00000-of-00002.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_4_small/data/train-00000-of-00002.parquet", + "identity": "d46c1712c79a2e72bbdbf5c463fed9bee62cc8271fd299a6f50dee94b789e14c", + "size": 539696180, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_4_small/data/train-00000-of-00002.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 45948, + "row_groups": 5, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 4, + "latency_ms": 133.33333333333334, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 400000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_5", + "source_revision": "67876ea9a8d5c687b924199574e298c68bf9b83d", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_5/data/train-00000-of-00001.parquet", + "identity": "0441d07d8b97c9ccf53b770c098caa3b999167d935997ec24685d1cdf5b4d56b", + "size": 183887566, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_5/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 261243, + "row_groups": 3, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 5, + "action_text": "LEFTFIRE", + "latency": 5, + "latency_ms": 333.33333333333337, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 500000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_7", + "source_revision": "d76a57ef1b84c591c47975347c539da59670f3b4", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_7/data/val-00000-of-00001.parquet", + "identity": "51047e60d6c8ec87f24b97c669f5affa51a7c585ee86f872577d179448f8cb9c", + "size": 40839139, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_7/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 58017, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 7, + "latency_ms": 466.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 11, + "t": 0, + "seed": 700011, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_0", + "source_revision": "6ee54a31fe5645bce299adeab17a68ff458bfabe", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_0/data/train-00000-of-00001.parquet", + "identity": "c7e5c9a1b249cce02b696b25d7069c094ff14d54b73da87b49a5b74f98115c03", + "size": 60683500, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_0/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 3836, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 22, + "action_text": "TURN_LEFT + MOVE_RIGHT", + "latency": 0, + "latency_ms": 0.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 0, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_8", + "source_revision": "164f4e56abb6c2637801e779918a99a7d70c9c3a", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_8/data/val-00000-of-00001.parquet", + "identity": "5d203d85aeea37d27cddd725e0fdbdafeefeb830f6ee1369164648377e7af524", + "size": 2375365, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_8/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 142, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 8, + "latency_ms": 228.57142857142858, + "reward": 0.0, + "done": false, + "episode_idx": 1, + "t": 0, + "seed": 800001, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_1_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_1_200ep/train.parquet", + "identity": "6c9d4241c9fc87b8f768a00a6550e7c525e337d7cf6fb0167b1e9f603467d716", + "size": 3409813276, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_1_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_1_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 297483, + "row_groups": 1163, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 1, + "action_text": "FLAP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 0, + "latency_ms": 33.333333333333336, + "latency_raw_frames": 1, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 1 raw frames (33.33 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 0, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_2_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_2_200ep/val.parquet", + "identity": "01b064893bb1eb3476095315a05ad13ef5a6eb96d706f658f97c5fce69df12c3", + "size": 822525480, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_2_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_2_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 72000, + "row_groups": 282, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 3, + "latency_ms": 66.66666666666667, + "latency_raw_frames": 2, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 2 raw frames (66.67 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 3, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_4_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_4_200ep/train.parquet", + "identity": "03966d3e348fe74eda6375c4e7b7e069dec265f947bfb9af5741f7cf0bf05b81", + "size": 3556756851, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_4_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_4_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 311588, + "row_groups": 1218, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 2, + "env_name": "flappy", + "episode_idx": 0, + "latency_ms": 133.33333333333334, + "latency_raw_frames": 4, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 4 raw frames (133.33 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 0, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_4", + "source_revision": "4dee6c0fd4abb86507001c4dc7404a5310c2f30a", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_4/data/val-00000-of-00001.parquet", + "identity": "3b898ef75b5f8ec49d2a5f2a8db0093d3205597b5232e515d41fbffafb15daa9", + "size": 4859123, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_4/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 291, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 4, + "latency_ms": 114.28571428571429, + "reward": 0.0, + "done": false, + "episode_idx": 10, + "t": 0, + "seed": 400010, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_3", + "source_revision": "b8b18213fe10aa0a6be2b7602e23c2ac938b49e0", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_3/data/train-00000-of-00001.parquet", + "identity": "32e0ba725e6c18b5696c1c761736503a5601c67ebc682b8bfba56e17d6f22179", + "size": 29029018, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_3/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 1735, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 27, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "latency": 3, + "latency_ms": 85.71428571428571, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 300000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_0", + "source_revision": "2deacb435543c52ae51f3c947bf917fdb0b5493c", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_0/data/val-00000-of-00001.parquet", + "identity": "85da89c53b147e5b69da3789829d554e9292a4f61232244dc870490207ff5b05", + "size": 25528820, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_0/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 36000, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 4, + "action_text": "RIGHTFIRE", + "latency": 0, + "latency_ms": 0.0, + "reward": 0.0, + "done": false, + "episode_idx": 2, + "t": 0, + "seed": 2, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_4", + "source_revision": "c7f6eba84f512bc22a095971c9b75499e841a026", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_4/data/train-00000-of-00001.parquet", + "identity": "2c344fbdb9bd89dd864b90860b4951efb39714f8822e1e20eaecf73c20a7b6fc", + "size": 188489295, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_4/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 263581, + "row_groups": 3, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 3, + "action_text": "LEFT", + "latency": 4, + "latency_ms": 266.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 400000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_14", + "source_revision": "fb78a1bc6280a4d0a6a60e8231d63d848819bd1e", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_14/data/val-00000-of-00001.parquet", + "identity": "7cb0a856d06b6d1e0ff53b995b7dfccc370111edaa2d221bcdf05a2150ec3de6", + "size": 3757224, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_14/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 238, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 28, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "latency": 14, + "latency_ms": 400.0, + "reward": 0.0, + "done": false, + "episode_idx": 24, + "t": 0, + "seed": 1400024, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_6", + "source_revision": "47cfced1713ce7d57d4567110269fb5929e8e7a8", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_6/data/train-00000-of-00001.parquet", + "identity": "fd9e8d9eaa7d40c230394130f66ce531c416bc767db8303acf15b8b4fc5eb421", + "size": 32877227, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_6/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2090, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 6, + "latency_ms": 171.42857142857142, + "reward": 0.0, + "done": false, + "episode_idx": 1, + "t": 0, + "seed": 600001, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "f90d1dac8b31a18c0496884729c744d7a4c877ae", + "source_path": "train.parquet", + "target_repo": "latency-sensitive-bench/memory-data", + "type": "dataset", + "path": "training/flappy_fixed_latency_3_200ep_7k2steps_ghost15/train.parquet", + "identity": "37d9e4a63c6cd77d68ebb8e4fee532fc081afadd209a36d43960b135988f0423", + "size": 25681178862, + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/memory-data", + "path": "training/flappy_fixed_latency_3_200ep_7k2steps_ghost15/train.parquet", + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd" + } + ], + "parquet_rows": 1396215, + "row_groups": 689, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 0, + "latency_ms": 100.0, + "latency_raw_frames": 3, + "raw_reward": 0.10000000149011612, + "seed": 0, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_11", + "source_revision": "78e4e1b6d9bb57c5f30712cd0ed2f3b02877ff56", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_11/data/val-00000-of-00001.parquet", + "identity": "5cab9f4b525d38604994fd39a905c8ac9095d770efe458c5420f61ab20d16880", + "size": 17613909, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_11/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 25428, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 11, + "latency_ms": 733.3333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 6, + "t": 0, + "seed": 1100006, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_7", + "source_revision": "d04b9bc42bfbb7da400438a9fb75c49b198f7776", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_7/data/train-00000-of-00001.parquet", + "identity": "4a362105927bd42632e69a09fca23b289f9607b2a51c15a70bbef77be492b08b", + "size": 35135665, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_7/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2157, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 7, + "latency_ms": 200.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 700000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_0_100ep_1ksteps_og/val.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_0_100ep_1ksteps_og/val.parquet", + "identity": "f2c758108f1b0f93eba0f6e7d57df0f413e23d8ffe89f2c7296aad69cd6de554", + "size": 31171805, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_0_100ep_1ksteps_og/val.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 10000, + "row_groups": 40, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 3, + "action_text": "push_right_soft", + "decision_step": 0, + "env_name": "inverted_pendulum_rgb_state", + "episode_idx": 1, + "latency_ms": 0.0, + "latency_raw_frames": 0, + "raw_reward": 1.0, + "seed": 1, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_2_1000ep_7k2steps/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/val-00000-of-00001.parquet", + "identity": "99a343f7af187728a2cd4d3fac69ea4e51157551cfe51b42aaf4a920a356198c", + "size": 5826746, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/val-00000-of-00001.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + }, + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/val-00000-of-00001.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 944, + "row_groups": 1, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 16, + "decision_step": 0, + "split": "val", + "seed": 42, + "latency_raw_frames": 2, + "latency_ms": 80.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_0_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_0_200ep/train.parquet", + "identity": "fcceb2dd4cfc7c2a088142612533e5bdbf65e61067f20e31928a0e30ab2d0958", + "size": 256157082, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_0_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_0_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 435000, + "row_groups": 1700, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 5, + "action_text": "LEFTFIRE", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 0, + "latency_ms": 0.0, + "latency_raw_frames": 0, + "raw_reward": 0.0, + "seed": 0, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_2_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_2_200ep/val.parquet", + "identity": "78a1a4c69f601d9333ae092f69254233d66c48f39b804d0fbed619926f45771f", + "size": 30312545, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_2_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_2_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 52326, + "row_groups": 205, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 2, + "action_text": "RIGHT", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 3, + "latency_ms": 133.33333333333334, + "latency_raw_frames": 8, + "raw_reward": 0.0, + "seed": 8, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_6_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_6_200ep/train.parquet", + "identity": "191ed66ae8a15c7767a507e474311176d34cd78fbd14dc4edd5307296b7a005a", + "size": 176346244, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_6_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_6_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 310953, + "row_groups": 1215, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 1, + "action_text": "FIRE", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 0, + "latency_ms": 400.0, + "latency_raw_frames": 24, + "raw_reward": 0.0, + "seed": 32, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_path": "demon_attack_fix_latency_8_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_8_200ep/val.parquet", + "identity": "59080949006a7db9d6fd11affe01ca21ccd857b79ca9fd48d073e10015a1c334", + "size": 22559655, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_200ep/demon_attack_fix_latency_8_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/demon_attack_200ep/0481972d94f6409a392c5b035f9a7f5c2681a96d/demon_attack_fix_latency_8_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 40599, + "row_groups": 159, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 1, + "action_text": "FIRE", + "decision_step": 0, + "env_name": "demon_attack", + "episode_idx": 3, + "latency_ms": 533.3333333333334, + "latency_raw_frames": 32, + "raw_reward": 0.0, + "seed": 27, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_1", + "source_revision": "e09b7f2c51726bb169d827fe973b2d429b6b1432", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_1/data/train-00000-of-00001.parquet", + "identity": "893c4234419bcce925c4e739d97c20b411fb1bf2cbbf30d085009195cefe86a2", + "size": 32981127, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_1/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 1966, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 29, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "latency": 1, + "latency_ms": 28.571428571428573, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 100000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_9", + "source_revision": "3405ac7c12ca266e2d3e83047bbd06e1e5eed220", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_9/data/val-00000-of-00001.parquet", + "identity": "b361a38c6d8838096be332baafb8dac29d3c54ecd10315842981b6ec39990981", + "size": 3297161, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_9/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 199, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 9, + "latency_ms": 257.14285714285717, + "reward": 0.0, + "done": false, + "episode_idx": 3, + "t": 0, + "seed": 900003, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_2_small", + "source_revision": "c9bbd1744020a587b55590c897628510e0741136", + "source_path": "data/train-00002-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_2_small/data/train-00002-of-00003.parquet", + "identity": "7fdce857810d7aff25aae41e3e1e825b8e380b3f7529605133e8efa91e74d38d", + "size": 381615970, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_2_small/data/train-00002-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 2, + "latency_ms": 66.66666666666667, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 2, + "t": 2400, + "seed": 200002, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_13", + "source_revision": "dbb7678df6396735290099d070dfb5827a4cf4a4", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_13/data/val-00000-of-00001.parquet", + "identity": "bf22ceaf3817f67308a7ade800567aed5b341c06b9c1255f63afd2534b937f35", + "size": 4381335, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_13/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 268, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 13, + "latency_ms": 371.42857142857144, + "reward": 0.0, + "done": false, + "episode_idx": 9, + "t": 0, + "seed": 1300009, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00002-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00002-of-00021.parquet", + "identity": "01ee577bd690f4f9d5278ee00858939847add716c4c3d63a72ff5721a7aa84fb", + "size": 457599942, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00002-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 18, + "decision_step": 6172, + "split": "train", + "seed": 28, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00005-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00005-of-00021.parquet", + "identity": "53ec0feeb2dd50ba4c6db9a0986acd3935fbf445a308c75f461cac36fa9fa012", + "size": 456627627, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00005-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 47, + "decision_step": 1030, + "split": "train", + "seed": 57, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00008-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00008-of-00021.parquet", + "identity": "2f483668bdec0fccec230c8e83b6891d57362afbf21db42f45cb10d4b44cf441", + "size": 456868082, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00008-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 75, + "decision_step": 3088, + "split": "train", + "seed": 85, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00011-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00011-of-00021.parquet", + "identity": "a6a5c03998dcdeb72d2ba50611595b4ae9ddf40114048bd4a2426d44098b8536", + "size": 456233963, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00011-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 103, + "decision_step": 5146, + "split": "train", + "seed": 113, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00014-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00014-of-00021.parquet", + "identity": "d9a4ebe2ca56b8f72cf235cc472a2c9a26f7caca9640b13838adf33fefdf3646", + "size": 455299955, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00014-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67886, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 133, + "decision_step": 4, + "split": "train", + "seed": 143, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00017-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00017-of-00021.parquet", + "identity": "8502bcff542d94e26f5945a814be763783997e32578a51970bedf4528d83da36", + "size": 455027310, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00017-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67885, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 161, + "decision_step": 2060, + "split": "train", + "seed": 171, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_path": "inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00020-of-00021.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00020-of-00021.parquet", + "identity": "34ffa01e7a6210ba62b7dfc26cc749ff84129af6e16f325c2100fd369fc08b46", + "size": 454340240, + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_0_200ep_7k2steps/train-00020-of-00021.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 67885, + "row_groups": 5, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 190, + "decision_step": 4115, + "split": "train", + "seed": 200, + "latency_raw_frames": 0, + "latency_ms": 0.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_3_small", + "source_revision": "8470ae1f00ca10bb8646ac7cfdccf9682e053eab", + "source_path": "data/train-00001-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_3_small/data/train-00001-of-00003.parquet", + "identity": "e7d9239375fd17cbd580a888bed0ab17ec8ce698bc89d7079098f34783f2b8d6", + "size": 382712552, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_3_small/data/train-00001-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 3, + "latency_ms": 100.0, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 1200, + "seed": 300000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_13", + "source_revision": "2a5c5d2aa4730db22fd33f4f00f4d93fc5043465", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_13/data/train-00000-of-00001.parquet", + "identity": "727a101510009e87037a76e3b8b647537d87cc476c347679a18bb174e49b4675", + "size": 105759043, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_13/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 154669, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 13, + "latency_ms": 866.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1300000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_2", + "source_revision": "c0605fa1df28b19249849c3e748bbc445bef6526", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_2/data/val-00000-of-00001.parquet", + "identity": "4754e38f5bd0e1b2cd720faa13c0688dd217fd84ba73c6ceb9bca6a73eb26f26", + "size": 51575172, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_2/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 72000, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 2, + "action_text": "RIGHT", + "latency": 2, + "latency_ms": 133.33333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 200000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_15", + "source_revision": "21811c6039f0a538f11821730ee6792bd0d6c0b4", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_15/data/train-00000-of-00001.parquet", + "identity": "77fd0b0f2fa68f8507491ac08ba7b67b0dd066a1a7db793ce38683230a4f32ba", + "size": 89408015, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_15/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 128892, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 1, + "action_text": "FIRE", + "latency": 15, + "latency_ms": 1000.0000000000001, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1500000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_9", + "source_revision": "7be93d38340141e385913e540b028024d74f1cf8", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_9/data/val-00000-of-00001.parquet", + "identity": "c237703705770820920c3ee8297ada52061fc5eee0ca74718173e7f88d7fc596", + "size": 28142328, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_9/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 41302, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 9, + "latency_ms": 600.0, + "reward": 0.0, + "done": false, + "episode_idx": 3, + "t": 0, + "seed": 900003, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_0_small", + "source_revision": "7e974f1e8e1ad7ca4ee0ffad0849d886d64128d3", + "source_path": "data/train-00000-of-00002.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_0_small/data/train-00000-of-00002.parquet", + "identity": "f4af8c1a68cd893b1cb6bd9cbdc124215e891b598d8806ccfac5f0b4f1be976f", + "size": 551074820, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_0_small/data/train-00000-of-00002.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 46943, + "row_groups": 5, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 0, + "latency_ms": 0.0, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 0, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_10", + "source_revision": "94762ebd79edb55d6e83ba2046729e3e6574d64d", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_10/data/train-00000-of-00001.parquet", + "identity": "be218fed0bb91c69feee31e7eb1ca2a3b6e0c15b3d96f918da26c5b73e455442", + "size": 32185015, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_10/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 1989, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 10, + "latency_ms": 285.7142857142857, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1000000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_1", + "source_revision": "7190bf45dfface92ec7cbaa3c774a7e29616db52", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_1/data/val-00000-of-00001.parquet", + "identity": "8859e2315a508e552b15e1838949890042fd371041e1188134529facc8687adb", + "size": 51361579, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_1/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 72000, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 2, + "action_text": "RIGHT", + "latency": 1, + "latency_ms": 66.66666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 10, + "t": 0, + "seed": 100010, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_8", + "source_revision": "e02c6b1db39b37965569d4f559f1d710fb55c58a", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_8/data/train-00000-of-00001.parquet", + "identity": "b2c69d835d97d644be310f80318168bc8968745ad537e10f1789446db156d713", + "size": 115935834, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_8/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 167228, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 1, + "action_text": "FIRE", + "latency": 8, + "latency_ms": 533.3333333333334, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 800000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_15", + "source_revision": "c586ddd84f76b6295de7929b62ad4da3359b0348", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_15/data/val-00000-of-00001.parquet", + "identity": "189425be1180cf29ede39f47929b5ae50d011f04819c251f33fe72a127c45079", + "size": 4005443, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_15/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 259, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 15, + "latency_ms": 428.57142857142856, + "reward": 0.0, + "done": false, + "episode_idx": 11, + "t": 0, + "seed": 1500011, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_1_small", + "source_revision": "749a849911b8c35005c26241fb4edbcee1286832", + "source_path": "data/train-00002-of-00003.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_1_small/data/train-00002-of-00003.parquet", + "identity": "73ab64af1ff56197361a9740156801d728723d7d2ece6406f4e2f50756747dd8", + "size": 379423798, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_1_small/data/train-00002-of-00003.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 32400, + "row_groups": 4, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 1, + "latency_ms": 33.333333333333336, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 0, + "t": 2400, + "seed": 100000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fix_latency_4_small", + "source_revision": "8a6a091570530476b313976b6a8d1b94d600f6d9", + "source_path": "data/train-00001-of-00002.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_fix_latency_4_small/data/train-00001-of-00002.parquet", + "identity": "7354873ef5be0870d9ef872c47a50de2b5461ff5fb0995e62392befdccdb50af", + "size": 543552440, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_fix_latency_4_small/data/train-00001-of-00002.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 45948, + "row_groups": 5, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "action_id": 0, + "action_text": "NOOP", + "latency": 4, + "latency_ms": 133.33333333333334, + "reward": 0.10000000149011612, + "done": false, + "episode_idx": 3, + "t": 1762, + "seed": 400003, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_5", + "source_revision": "67876ea9a8d5c687b924199574e298c68bf9b83d", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_5/data/val-00000-of-00001.parquet", + "identity": "b3ee77c442f6fda3a8a682e7adfbaef01636c6ea65afef9f249d6bd56b63eb3f", + "size": 48472055, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_5/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 68636, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 5, + "action_text": "LEFTFIRE", + "latency": 5, + "latency_ms": 333.33333333333337, + "reward": 0.0, + "done": false, + "episode_idx": 10, + "t": 0, + "seed": 500010, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_10", + "source_revision": "9a3f7f736402b063885068e683860b6746e5337d", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_10/data/train-00000-of-00001.parquet", + "identity": "4edf6cf1be95ac4c739bddc5cbfb963cea4c9908eac3f6b4b0a25f9abeff2dfd", + "size": 101430465, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_10/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 145281, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 10, + "latency_ms": 666.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1000000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_0", + "source_revision": "6ee54a31fe5645bce299adeab17a68ff458bfabe", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_0/data/val-00000-of-00001.parquet", + "identity": "0931d75b4b291662663d0703150915e6f7e60408c7e836e7dc3358df8a7d3a6b", + "size": 7059124, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_0/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 443, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 22, + "action_text": "TURN_LEFT + MOVE_RIGHT", + "latency": 0, + "latency_ms": 0.0, + "reward": 0.0, + "done": false, + "episode_idx": 2, + "t": 0, + "seed": 2, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_0_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_0_200ep/train.parquet", + "identity": "db4ce27834da008d1e0ac594b468b74f246699899d081207e3d66383ea1304bb", + "size": 3213015482, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_0_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_0_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 279882, + "row_groups": 1094, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 2, + "env_name": "flappy", + "episode_idx": 0, + "latency_ms": 0.0, + "latency_raw_frames": 0, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 0 raw frames (0.00 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 0, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_1_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_1_200ep/val.parquet", + "identity": "12a164c7536122404d78675a3aa7f3990036d6d3383982394460598e9d53698a", + "size": 822260231, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_1_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_1_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 72000, + "row_groups": 282, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 1, + "action_text": "FLAP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 3, + "latency_ms": 33.333333333333336, + "latency_raw_frames": 1, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 1 raw frames (33.33 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 3, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_3_200ep/train.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_3_200ep/train.parquet", + "identity": "f94ff88ce70cbcd17ad5c1ac5d17e523f1eba7e012d55518973000d61861fbde", + "size": 3610054061, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_3_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_3_200ep/train.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 316528, + "row_groups": 1237, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 20, + "env_name": "flappy", + "episode_idx": 0, + "latency_ms": 100.0, + "latency_raw_frames": 3, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 3 raw frames (100.00 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 518, + "split": "train" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_path": "flappy_fix_latency_4_200ep/val.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/flappy_200ep/flappy_fix_latency_4_200ep/val.parquet", + "identity": "adf760bcac9d4e50620bffcb3202e806ff9da39be7f12d57cbb098ed205b6b06", + "size": 819988085, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/flappy_200ep/flappy_fix_latency_4_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + }, + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training_versions/flappy_200ep/5df5688a961dc2c91090155c719bff4a5b248235/flappy_fix_latency_4_200ep/val.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 72000, + "row_groups": 282, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 3, + "latency_ms": 133.33333333333334, + "latency_raw_frames": 4, + "prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP. Current action latency is 4 raw frames (133.33 ms). The environment runs at 30 FPS and observations are emitted at 30 FPS. Choose the best next action.", + "raw_reward": 0.10000000149011612, + "seed": 3, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_12", + "source_revision": "ac9d21a59b908e717de2bee2cc749b90495a00b8", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_12/data/train-00000-of-00001.parquet", + "identity": "81506b959eb573139b037c20da50b2f270a3afccbc728dacebec99281a27f18e", + "size": 116748129, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_12/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 169767, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 0, + "action_text": "NOOP", + "latency": 12, + "latency_ms": 800.0, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 1200000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_3", + "source_revision": "b8b18213fe10aa0a6be2b7602e23c2ac938b49e0", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_3/data/val-00000-of-00001.parquet", + "identity": "edec5da7a8c6689d14f7c065beff41cec01a3075fc4ab15eb82d3061c3408d64", + "size": 2944071, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_3/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 174, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 27, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "latency": 3, + "latency_ms": 85.71428571428571, + "reward": 0.0, + "done": false, + "episode_idx": 6, + "t": 0, + "seed": 300006, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_5", + "source_revision": "f614a91258d86cb34cd2e3e0e79e2bf69b321e76", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_5/data/train-00000-of-00001.parquet", + "identity": "a3d2e5475853c420c3659e733248e53fb5406c79e618723c8b5c5a32eefc40e3", + "size": 35923917, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_5/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2139, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 24, + "action_text": "TURN_LEFT + MOVE_FORWARD", + "latency": 5, + "latency_ms": 142.85714285714286, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 500000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_4", + "source_revision": "c7f6eba84f512bc22a095971c9b75499e841a026", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_4/data/val-00000-of-00001.parquet", + "identity": "c2bb90a723e6cf72d378793f8a20bdc95940105220bbfa70ef77616c835a03e5", + "size": 46025266, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_4/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 64664, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 3, + "action_text": "LEFT", + "latency": 4, + "latency_ms": 266.6666666666667, + "reward": 0.0, + "done": false, + "episode_idx": 2, + "t": 0, + "seed": 400002, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/demon_attack_fix_latency_6", + "source_revision": "0766ba4ea9196b861302e99e564b7d2e05453885", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/demon_attack_fix_latency_6/data/train-00000-of-00001.parquet", + "identity": "b6d4c64ebc3ef04a1ffadd75b4188d815f3ef80d894ac7042fbd2e241fd2634b", + "size": 159586439, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/demon_attack_fix_latency_6/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 226141, + "row_groups": 2, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "prompt": "You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE. Survive, avoid enemy fire, and shoot enemies to maximize score.", + "action_id": 3, + "action_text": "LEFT", + "latency": 6, + "latency_ms": 400.0, + "reward": 0.0, + "done": false, + "episode_idx": 1, + "t": 0, + "seed": 600001, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_6", + "source_revision": "47cfced1713ce7d57d4567110269fb5929e8e7a8", + "source_path": "data/val-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_6/data/val-00000-of-00001.parquet", + "identity": "ce7755f960102e952bb7d28db79bd1be1640a7f715826f6222a913e81966c5cd", + "size": 3187961, + "revision": "27031df0538043a2a2e73540d65329f96c82b98f", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_6/data/val-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 191, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "val", + "action_id": 26, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "latency": 6, + "latency_ms": 171.42857142857142, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 600000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + } +] \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_0_200ep_context5/filter_report.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_0_200ep_context5/filter_report.json new file mode 100644 index 0000000000000000000000000000000000000000..7d6734b6780842fdc876bcebace10930123c508a --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_0_200ep_context5/filter_report.json @@ -0,0 +1,12 @@ +{ + "accepted_episodes": 200, + "accepted_train_episodes": 180, + "accepted_val_episodes": 20, + "attempted_episodes": 227, + "preset": "clean_v1", + "rejected_episodes": 27, + "train_rows_after_row_filter": 279882, + "train_rows_before_row_filter": 648000, + "val_rows_after_row_filter": 72000, + "val_rows_before_row_filter": 72000 +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_0_200ep_context5/metadata.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_0_200ep_context5/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..55577b7ef30179726cd55a35078e8deaf09c7652 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_0_200ep_context5/metadata.json @@ -0,0 +1,305 @@ +{ + "action_labels": [ + "NOOP", + "FLAP" + ], + "action_layout": "flappy_discrete", + "base_prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "checkpoint_config": { + "actor_critic_share_weights": true, + "actor_worker_gpus": [ + 0 + ], + "adam_beta1": 0.9, + "adam_beta2": 0.999, + "adam_eps": 1e-05, + "adaptive_stddev": true, + "add_latency_info": false, + "algo": "APPO", + "async_rl": true, + "audio_on": false, + "background": "day", + "batch_size": 4096, + "batched_sampling": true, + "benchmark": false, + "bird_color": "yellow", + "cli_args": { + "actor_worker_gpus": [ + 0 + ], + "add_latency_info": false, + "algo": "APPO", + "async_rl": true, + "audio_on": false, + "background": "day", + "batch_size": 4096, + "batched_sampling": true, + "bird_color": "yellow", + "debug": false, + "debug_timelimit_diagnostics": false, + "device": "gpu", + "env": "latency_flappy", + "env_gpu_actions": false, + "env_gpu_observations": true, + "episode_metrics_path": "results/checkpoints/flappy_sf_same_latency_l0_seed0/episode_metrics.jsonl", + "episodes_per_stage": 10, + "eval_deterministic": true, + "eval_episodes": 20, + "eval_max_steps": 3600, + "eval_queue_latency_values": "0,1,2,3,4,5", + "experiment": "flappy_sf_same_latency_l0_seed0", + "experiment_summaries_interval": 1, + "frame_stack": 4, + "gae_lambda": 0.95, + "gamma": 0.99, + "gpu_render_batch_size": 64, + "gpu_render_device": "auto", + "gpu_render_profile": false, + "gpu_render_profile_interval": 200, + "gym_id": "FlappyBird-v0", + "keep_checkpoints": 5, + "latency_mode": "episode_fixed", + "latency_sampling_mode": "uniform_all", + "learning_rate": 0.00025, + "max_grad_norm": 0.5, + "mode": "train", + "normalize_input": true, + "normalize_obs": true, + "normalize_returns": true, + "num_batches_per_epoch": 8, + "num_envs_per_worker": 1, + "num_epochs": 8, + "num_policies": 1, + "num_workers": 2, + "obs_mode": "image_only", + "obs_resize": "84,84", + "pipe_color": "green", + "pipe_gap": 100, + "ppo_clip_ratio": 0.1, + "ppo_clip_value": 0.2, + "queue_latency_values": "0", + "randomize_within_stage": false, + "recurrence": 1, + "restart_behavior": "resume", + "rollout": 256, + "save_every_sec": 600, + "seed": 0, + "stats_avg": 100, + "train_dir": "results/checkpoints", + "train_for_env_steps": 15000000, + "use_gpu_render": true, + "use_lidar": false, + "use_rnn": false, + "value_loss_coeff": 0.5, + "wandb_group": "flappy-new(remove training episode limit)", + "wandb_job_type": "sample_factory", + "wandb_project": "latency-sensitive-bench", + "with_wandb": true, + "within_episode_change_interval": 1, + "within_episode_change_prob": 1.0, + "worker_num_splits": 1 + }, + "command_line": "--mode train --algo APPO --env latency_flappy --experiment flappy_sf_same_latency_l0_seed0 --train_dir results/checkpoints --restart_behavior resume --device gpu --actor_worker_gpus 0 --env_gpu_observations True --env_gpu_actions False --gpu-render-batch-size 64 --seed 0 --episode_metrics_path results/checkpoints/flappy_sf_same_latency_l0_seed0/episode_metrics.jsonl --train_for_env_steps 15000000 --num_workers 2 --num_envs_per_worker 1 --num_policies 1 --batch_size 4096 --rollout 256 --recurrence 1 --num_epochs 8 --num_batches_per_epoch 8 --worker_num_splits 1 --learning_rate 0.00025 --gamma 0.99 --gae_lambda 0.95 --ppo_clip_ratio 0.1 --ppo_clip_value 0.2 --value_loss_coeff 0.5 --max_grad_norm 0.5 --save_every_sec 600 --keep_checkpoints 5 --stats_avg 100 --experiment_summaries_interval 1 --batched_sampling True --async_rl True --use_rnn False --normalize_returns True --normalize_input True --queue-latency-values 0 --latency-mode episode_fixed --latency-sampling-mode uniform_all --episodes-per-stage 10 --randomize-within-stage False --within-episode-change-interval 1 --within-episode-change-prob 1.0 --add-latency-info False --eval-episodes 20 --eval-queue-latency-values 0,1,2,3,4,5 --eval-max-steps 3600 --eval-deterministic True --with_wandb True --wandb_project latency-sensitive-bench --wandb_group flappy-new(remove training episode limit) --wandb_job_type sample_factory --gym_id FlappyBird-v0 --use_lidar False --normalize_obs True --audio_on False --obs_resize 84,84 --obs_mode image_only --use-gpu-render True --gpu-render-device auto --gpu-render-profile False --gpu-render-profile-interval 200 --pipe_gap 100 --bird_color yellow --pipe_color green --background day --frame_stack 4 --debug False --debug-timelimit-diagnostics False", + "continuous_tanh_scale": 0.0, + "debug": false, + "debug_timelimit_diagnostics": false, + "decoder_mlp_layers": [], + "decorrelate_envs_on_one_worker": true, + "decorrelate_experience_max_seconds": 0, + "default_niceness": 0, + "device": "gpu", + "encoder_conv_architecture": "convnet_atari", + "encoder_conv_mlp_layers": [ + 512 + ], + "encoder_mlp_layers": [ + 512, + 512 + ], + "env": "latency_flappy", + "env_frameskip": 1, + "env_framestack": 1, + "env_gpu_actions": false, + "env_gpu_observations": true, + "episode_metrics_path": "results/checkpoints/flappy_sf_same_latency_l0_seed0/episode_metrics.jsonl", + "episodes_per_stage": 10, + "eval_deterministic": true, + "eval_episodes": 20, + "eval_max_steps": 3600, + "eval_queue_latency_values": "0,1,2,3,4,5", + "eval_raw_reward": false, + "experiment": "flappy_sf_same_latency_l0_seed0", + "experiment_summaries_interval": 1, + "exploration_loss": "entropy", + "exploration_loss_coeff": 0.003, + "flush_summaries_interval": 30, + "force_envs_single_thread": false, + "frame_stack": 4, + "gae_lambda": 0.95, + "gamma": 0.99, + "git_hash": "8b374967e3be2fdc51a3469ce7955c63e6cfa046", + "git_repo_name": "git@github.com:ZihanWang314/latency-sensitive-bench.git", + "gpu_render_batch_size": 64, + "gpu_render_device": "auto", + "gpu_render_profile": false, + "gpu_render_profile_interval": 200, + "gym_id": "FlappyBird-v0", + "heartbeat_interval": 20, + "heartbeat_reporting_interval": 180, + "help": false, + "initial_stddev": 1.0, + "keep_checkpoints": 5, + "kl_loss_coeff": 0.0, + "latency_mode": "episode_fixed", + "latency_sampling_mode": "uniform_all", + "learning_rate": 0.00025, + "load_checkpoint_kind": "latest", + "log_to_file": true, + "lr_adaptive_max": 0.01, + "lr_adaptive_min": 1e-06, + "lr_schedule": "linear_decay", + "lr_schedule_kl_threshold": 0.008, + "max_episode_steps": 0, + "max_grad_norm": 0.5, + "max_policy_lag": 1000, + "mode": "train", + "nonlinearity": "elu", + "normalize_input": true, + "normalize_input_keys": null, + "normalize_obs": true, + "normalize_returns": true, + "num_batches_per_epoch": 8, + "num_batches_to_accumulate": 2, + "num_envs_per_worker": 1, + "num_epochs": 8, + "num_policies": 1, + "num_workers": 2, + "obs_mode": "image_only", + "obs_resize": "84,84", + "obs_scale": 255.0, + "obs_subtract_mean": 0.0, + "optimizer": "adam", + "output_dir": "outputs/train/flappy/latency_0/seed_0", + "pbt_mix_policies_in_one_env": true, + "pbt_mutation_rate": 0.15, + "pbt_optimize_gamma": false, + "pbt_period_env_steps": 5000000, + "pbt_perturb_max": 1.5, + "pbt_perturb_min": 1.1, + "pbt_replace_fraction": 0.3, + "pbt_replace_reward_gap": 0.1, + "pbt_replace_reward_gap_absolute": 1e-06, + "pbt_start_mutation": 20000000, + "pbt_target_objective": "true_objective", + "pipe_color": "green", + "pipe_gap": 100, + "pixel_format": "CHW", + "policy_init_gain": 1.0, + "policy_initialization": "orthogonal", + "policy_workers_per_policy": 1, + "ppo_clip_ratio": 0.1, + "ppo_clip_value": 0.2, + "queue_latency_values": "0", + "randomize_within_stage": false, + "recurrence": 1, + "restart_behavior": "resume", + "reward_clip": 1000.0, + "reward_scale": 1.0, + "rnn_num_layers": 1, + "rnn_size": 512, + "rnn_type": "gru", + "rollout": 256, + "save_best_after": 100000, + "save_best_every_sec": 5, + "save_best_metric": "reward", + "save_every_sec": 600, + "save_milestones_sec": -1, + "score_limit": -1, + "screen_size": "", + "seed": 0, + "serial_mode": false, + "set_workers_cpu_affinity": true, + "shuffle_minibatches": false, + "stats_avg": 100, + "summaries_use_frameskip": true, + "train_dir": "results/checkpoints", + "train_for_env_steps": 15000000, + "train_for_seconds": 10000000000, + "use_env_info_cache": false, + "use_gpu_render": true, + "use_lidar": false, + "use_record_episode_statistics": false, + "use_rnn": false, + "value_bootstrap": false, + "value_loss_coeff": 0.5, + "vtrace_c": 1.0, + "vtrace_rho": 1.0, + "wandb_group": "flappy-new(remove training episode limit)", + "wandb_job_type": "sample_factory", + "wandb_project": "latency-sensitive-bench", + "wandb_tags": [], + "wandb_unique_id": "flappy-bird-l0-s0", + "wandb_user": null, + "with_pbt": false, + "with_vtrace": false, + "with_wandb": true, + "within_episode_change_interval": 1, + "within_episode_change_prob": 1.0, + "worker_num_splits": 1 + }, + "checkpoint_experiment_dir": "/workspace/latency-sensitive-bench/checkpoints/flappy_sf_same_latency_l0_seed0", + "checkpoint_path": null, + "checkpoint_train_step": 29312, + "config_source": "/workspace/latency-sensitive-bench/checkpoints/flappy_sf_same_latency_l0_seed0/config.json", + "context_image_count": 4, + "context_window": 5, + "env_fps": 30.0, + "env_name": "flappy", + "episode_return_field": "episode_raw_return", + "filter_preset": "clean_v1", + "filter_report": { + "accepted_episodes": 200, + "accepted_train_episodes": 180, + "accepted_val_episodes": 20, + "attempted_episodes": 227, + "preset": "clean_v1", + "rejected_episodes": 27, + "train_rows_after_row_filter": 279882, + "train_rows_before_row_filter": 648000, + "val_rows_after_row_filter": 72000, + "val_rows_before_row_filter": 72000 + }, + "frame_stack": 4, + "image_resolution": [ + 512, + 288, + 3 + ], + "image_source": "env_raw_rgb", + "latency_override": { + "add_latency_info": false, + "fixed_latency_ms": 0.0, + "type": "fixed" + }, + "latency_source": "legacy_checkpoint_name", + "max_episodes": 200, + "max_steps_per_episode": 3600, + "num_envs": 2048, + "obs_fps": 30.0, + "obs_stride_raw_frames": 1, + "prompt_source": "latency_bench.data.prompts", + "prompt_timing_suffix": " Current action latency is {latency_raw_frames} raw frames ({latency_ms:.2f} ms). The environment runs at {env_fps:g} FPS and observations are emitted at {obs_fps:g} FPS. Choose the best next action.", + "replay_num_envs": 16, + "reward_field": "raw_reward", + "reward_semantics": "raw_reward", + "rows_unit": "decision_step", + "schema_version": 2, + "split_seed": 0, + "total_episodes": 200, + "total_rows": 351882, + "train_episodes": 180, + "train_rows": 279882, + "val_episodes": 20, + "val_fraction": 0.1, + "val_rows": 72000 +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_1_200ep_context5/filter_report.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_1_200ep_context5/filter_report.json new file mode 100644 index 0000000000000000000000000000000000000000..87f157c1b60a0d25fe6dda16245f039a588ead78 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_1_200ep_context5/filter_report.json @@ -0,0 +1,12 @@ +{ + "accepted_episodes": 200, + "accepted_train_episodes": 180, + "accepted_val_episodes": 20, + "attempted_episodes": 200, + "preset": "clean_v1", + "rejected_episodes": 0, + "train_rows_after_row_filter": 297483, + "train_rows_before_row_filter": 648000, + "val_rows_after_row_filter": 72000, + "val_rows_before_row_filter": 72000 +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_1_200ep_context5/metadata.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_1_200ep_context5/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..6bd8f76284eaa0ea54eb0503b251ea7b5d2c0ce9 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/flappy_fix_latency_1_200ep_context5/metadata.json @@ -0,0 +1,305 @@ +{ + "action_labels": [ + "NOOP", + "FLAP" + ], + "action_layout": "flappy_discrete", + "base_prompt": "You are playing Flappy Bird. Pass through the pipe gaps and stay alive. Choose the action: NOOP, FLAP.", + "checkpoint_config": { + "actor_critic_share_weights": true, + "actor_worker_gpus": [ + 0 + ], + "adam_beta1": 0.9, + "adam_beta2": 0.999, + "adam_eps": 1e-05, + "adaptive_stddev": true, + "add_latency_info": false, + "algo": "APPO", + "async_rl": true, + "audio_on": false, + "background": "day", + "batch_size": 4096, + "batched_sampling": true, + "benchmark": false, + "bird_color": "yellow", + "cli_args": { + "actor_worker_gpus": [ + 0 + ], + "add_latency_info": false, + "algo": "APPO", + "async_rl": true, + "audio_on": false, + "background": "day", + "batch_size": 4096, + "batched_sampling": true, + "bird_color": "yellow", + "debug": false, + "debug_timelimit_diagnostics": false, + "device": "gpu", + "env": "latency_flappy", + "env_gpu_actions": false, + "env_gpu_observations": true, + "episode_metrics_path": "results/checkpoints/flappy_sf_same_latency_l1_seed0/episode_metrics.jsonl", + "episodes_per_stage": 10, + "eval_deterministic": true, + "eval_episodes": 20, + "eval_max_steps": 3600, + "eval_queue_latency_values": "0,1,2,3,4,5", + "experiment": "flappy_sf_same_latency_l1_seed0", + "experiment_summaries_interval": 1, + "frame_stack": 4, + "gae_lambda": 0.95, + "gamma": 0.99, + "gpu_render_batch_size": 64, + "gpu_render_device": "auto", + "gpu_render_profile": false, + "gpu_render_profile_interval": 200, + "gym_id": "FlappyBird-v0", + "keep_checkpoints": 5, + "latency_mode": "episode_fixed", + "latency_sampling_mode": "uniform_all", + "learning_rate": 0.00025, + "max_grad_norm": 0.5, + "mode": "train", + "normalize_input": true, + "normalize_obs": true, + "normalize_returns": true, + "num_batches_per_epoch": 8, + "num_envs_per_worker": 1, + "num_epochs": 8, + "num_policies": 1, + "num_workers": 2, + "obs_mode": "image_only", + "obs_resize": "84,84", + "pipe_color": "green", + "pipe_gap": 100, + "ppo_clip_ratio": 0.1, + "ppo_clip_value": 0.2, + "queue_latency_values": "1", + "randomize_within_stage": false, + "recurrence": 1, + "restart_behavior": "resume", + "rollout": 256, + "save_every_sec": 600, + "seed": 0, + "stats_avg": 100, + "train_dir": "results/checkpoints", + "train_for_env_steps": 15000000, + "use_gpu_render": true, + "use_lidar": false, + "use_rnn": false, + "value_loss_coeff": 0.5, + "wandb_group": "flappy-new(remove training episode limit)", + "wandb_job_type": "sample_factory", + "wandb_project": "latency-sensitive-bench", + "with_wandb": true, + "within_episode_change_interval": 1, + "within_episode_change_prob": 1.0, + "worker_num_splits": 1 + }, + "command_line": "--mode train --algo APPO --env latency_flappy --experiment flappy_sf_same_latency_l1_seed0 --train_dir results/checkpoints --restart_behavior resume --device gpu --actor_worker_gpus 0 --env_gpu_observations True --env_gpu_actions False --gpu-render-batch-size 64 --seed 0 --episode_metrics_path results/checkpoints/flappy_sf_same_latency_l1_seed0/episode_metrics.jsonl --train_for_env_steps 15000000 --num_workers 2 --num_envs_per_worker 1 --num_policies 1 --batch_size 4096 --rollout 256 --recurrence 1 --num_epochs 8 --num_batches_per_epoch 8 --worker_num_splits 1 --learning_rate 0.00025 --gamma 0.99 --gae_lambda 0.95 --ppo_clip_ratio 0.1 --ppo_clip_value 0.2 --value_loss_coeff 0.5 --max_grad_norm 0.5 --save_every_sec 600 --keep_checkpoints 5 --stats_avg 100 --experiment_summaries_interval 1 --batched_sampling True --async_rl True --use_rnn False --normalize_returns True --normalize_input True --queue-latency-values 1 --latency-mode episode_fixed --latency-sampling-mode uniform_all --episodes-per-stage 10 --randomize-within-stage False --within-episode-change-interval 1 --within-episode-change-prob 1.0 --add-latency-info False --eval-episodes 20 --eval-queue-latency-values 0,1,2,3,4,5 --eval-max-steps 3600 --eval-deterministic True --with_wandb True --wandb_project latency-sensitive-bench --wandb_group flappy-new(remove training episode limit) --wandb_job_type sample_factory --gym_id FlappyBird-v0 --use_lidar False --normalize_obs True --audio_on False --obs_resize 84,84 --obs_mode image_only --use-gpu-render True --gpu-render-device auto --gpu-render-profile False --gpu-render-profile-interval 200 --pipe_gap 100 --bird_color yellow --pipe_color green --background day --frame_stack 4 --debug False --debug-timelimit-diagnostics False", + "continuous_tanh_scale": 0.0, + "debug": false, + "debug_timelimit_diagnostics": false, + "decoder_mlp_layers": [], + "decorrelate_envs_on_one_worker": true, + "decorrelate_experience_max_seconds": 0, + "default_niceness": 0, + "device": "gpu", + "encoder_conv_architecture": "convnet_atari", + "encoder_conv_mlp_layers": [ + 512 + ], + "encoder_mlp_layers": [ + 512, + 512 + ], + "env": "latency_flappy", + "env_frameskip": 1, + "env_framestack": 1, + "env_gpu_actions": false, + "env_gpu_observations": true, + "episode_metrics_path": "results/checkpoints/flappy_sf_same_latency_l1_seed0/episode_metrics.jsonl", + "episodes_per_stage": 10, + "eval_deterministic": true, + "eval_episodes": 20, + "eval_max_steps": 3600, + "eval_queue_latency_values": "0,1,2,3,4,5", + "eval_raw_reward": false, + "experiment": "flappy_sf_same_latency_l1_seed0", + "experiment_summaries_interval": 1, + "exploration_loss": "entropy", + "exploration_loss_coeff": 0.003, + "flush_summaries_interval": 30, + "force_envs_single_thread": false, + "frame_stack": 4, + "gae_lambda": 0.95, + "gamma": 0.99, + "git_hash": "8b374967e3be2fdc51a3469ce7955c63e6cfa046", + "git_repo_name": "git@github.com:ZihanWang314/latency-sensitive-bench.git", + "gpu_render_batch_size": 64, + "gpu_render_device": "auto", + "gpu_render_profile": false, + "gpu_render_profile_interval": 200, + "gym_id": "FlappyBird-v0", + "heartbeat_interval": 20, + "heartbeat_reporting_interval": 180, + "help": false, + "initial_stddev": 1.0, + "keep_checkpoints": 5, + "kl_loss_coeff": 0.0, + "latency_mode": "episode_fixed", + "latency_sampling_mode": "uniform_all", + "learning_rate": 0.00025, + "load_checkpoint_kind": "latest", + "log_to_file": true, + "lr_adaptive_max": 0.01, + "lr_adaptive_min": 1e-06, + "lr_schedule": "linear_decay", + "lr_schedule_kl_threshold": 0.008, + "max_episode_steps": 0, + "max_grad_norm": 0.5, + "max_policy_lag": 1000, + "mode": "train", + "nonlinearity": "elu", + "normalize_input": true, + "normalize_input_keys": null, + "normalize_obs": true, + "normalize_returns": true, + "num_batches_per_epoch": 8, + "num_batches_to_accumulate": 2, + "num_envs_per_worker": 1, + "num_epochs": 8, + "num_policies": 1, + "num_workers": 2, + "obs_mode": "image_only", + "obs_resize": "84,84", + "obs_scale": 255.0, + "obs_subtract_mean": 0.0, + "optimizer": "adam", + "output_dir": "outputs/train/flappy/latency_1/seed_0", + "pbt_mix_policies_in_one_env": true, + "pbt_mutation_rate": 0.15, + "pbt_optimize_gamma": false, + "pbt_period_env_steps": 5000000, + "pbt_perturb_max": 1.5, + "pbt_perturb_min": 1.1, + "pbt_replace_fraction": 0.3, + "pbt_replace_reward_gap": 0.1, + "pbt_replace_reward_gap_absolute": 1e-06, + "pbt_start_mutation": 20000000, + "pbt_target_objective": "true_objective", + "pipe_color": "green", + "pipe_gap": 100, + "pixel_format": "CHW", + "policy_init_gain": 1.0, + "policy_initialization": "orthogonal", + "policy_workers_per_policy": 1, + "ppo_clip_ratio": 0.1, + "ppo_clip_value": 0.2, + "queue_latency_values": "1", + "randomize_within_stage": false, + "recurrence": 1, + "restart_behavior": "resume", + "reward_clip": 1000.0, + "reward_scale": 1.0, + "rnn_num_layers": 1, + "rnn_size": 512, + "rnn_type": "gru", + "rollout": 256, + "save_best_after": 100000, + "save_best_every_sec": 5, + "save_best_metric": "reward", + "save_every_sec": 600, + "save_milestones_sec": -1, + "score_limit": -1, + "screen_size": "", + "seed": 0, + "serial_mode": false, + "set_workers_cpu_affinity": true, + "shuffle_minibatches": false, + "stats_avg": 100, + "summaries_use_frameskip": true, + "train_dir": "results/checkpoints", + "train_for_env_steps": 15000000, + "train_for_seconds": 10000000000, + "use_env_info_cache": false, + "use_gpu_render": true, + "use_lidar": false, + "use_record_episode_statistics": false, + "use_rnn": false, + "value_bootstrap": false, + "value_loss_coeff": 0.5, + "vtrace_c": 1.0, + "vtrace_rho": 1.0, + "wandb_group": "flappy-new(remove training episode limit)", + "wandb_job_type": "sample_factory", + "wandb_project": "latency-sensitive-bench", + "wandb_tags": [], + "wandb_unique_id": "flappy-bird-l1-s0", + "wandb_user": null, + "with_pbt": false, + "with_vtrace": false, + "with_wandb": true, + "within_episode_change_interval": 1, + "within_episode_change_prob": 1.0, + "worker_num_splits": 1 + }, + "checkpoint_experiment_dir": "/workspace/latency-sensitive-bench/checkpoints/flappy_sf_same_latency_l1_seed0", + "checkpoint_path": null, + "checkpoint_train_step": 29328, + "config_source": "/workspace/latency-sensitive-bench/checkpoints/flappy_sf_same_latency_l1_seed0/config.json", + "context_image_count": 4, + "context_window": 5, + "env_fps": 30.0, + "env_name": "flappy", + "episode_return_field": "episode_raw_return", + "filter_preset": "clean_v1", + "filter_report": { + "accepted_episodes": 200, + "accepted_train_episodes": 180, + "accepted_val_episodes": 20, + "attempted_episodes": 200, + "preset": "clean_v1", + "rejected_episodes": 0, + "train_rows_after_row_filter": 297483, + "train_rows_before_row_filter": 648000, + "val_rows_after_row_filter": 72000, + "val_rows_before_row_filter": 72000 + }, + "frame_stack": 4, + "image_resolution": [ + 512, + 288, + 3 + ], + "image_source": "env_raw_rgb", + "latency_override": { + "add_latency_info": false, + "fixed_latency_ms": 33.333333333333336, + "type": "fixed" + }, + "latency_source": "legacy_checkpoint_name", + "max_episodes": 200, + "max_steps_per_episode": 3600, + "num_envs": 2048, + "obs_fps": 30.0, + "obs_stride_raw_frames": 1, + "prompt_source": "latency_bench.data.prompts", + "prompt_timing_suffix": " Current action latency is {latency_raw_frames} raw frames ({latency_ms:.2f} ms). The environment runs at {env_fps:g} FPS and observations are emitted at {obs_fps:g} FPS. Choose the best next action.", + "replay_num_envs": 16, + "reward_field": "raw_reward", + "reward_semantics": "raw_reward", + "rows_unit": "decision_step", + "schema_version": 2, + "split_seed": 0, + "total_episodes": 200, + "total_rows": 369483, + "train_episodes": 180, + "train_rows": 297483, + "val_episodes": 20, + "val_fraction": 0.1, + "val_rows": 72000 +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.full.yaml b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.full.yaml new file mode 100644 index 0000000000000000000000000000000000000000..2908cf70ca82e02c259934b4e4b22dfc0b5163c1 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.full.yaml @@ -0,0 +1,261 @@ +framework: + name: WanOFT + world_model: + base_wm: /workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + extract_layers: + - -1 + num_frames: 5 + qwenvl: + base_vlm: /workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + attn_implementation: sdpa + enable_gradient_checkpointing: true + action_model: + action_model_type: MLP + action_dim: 7 + action_hidden_dim: 3072 + action_horizon: 8 + future_action_window_size: 7 + past_action_window_size: 0 + loss_type: current_discrete_ce + class_weights: null + future_loss_weight: null + action_query_source: mean + action_query_num_heads: 24 + state_dim: 7 + action_env_dim: 2 +datasets: + vla_data: + dataset_py: lerobot_datasets + include_state: true + data_root_dir: /workspace/starVLA/data/flappy_fix_latency_1_200ep_context5 + data_mix: flappy_train__bridge + eval_data_mix: flappy_train__bridge__val + custom_mixtures_path: null + action_type: discrete + sequential_step_sampling: false + eval_sequential_step_sampling: null + num_workers: 8 + eval_num_workers: 8 + prefetch_factor: 4 + persistent_workers: true + pin_memory: true + shuffle: true + action_balance: + enabled: false + strategy: balanced_epoch + action_key: action_id + target_flap_fraction: 0.3 + noop_id: 0 + flap_id: 1 + latency_curriculum: + enabled: false + strategy: exclusive + latencies: null + phase_steps: null + new_latency_passes: 1.0 + replay_passes: 0.25 + target_total_passes: 2.0 + final_equalization: true + step_budget_mode: auto + eval_at_phase_end: false + save_at_phase_end: false + computed_plan: null + effective_batch_size: null + per_device_batch_size: 4 + load_all_data_for_training: true + num_obs_frames: 1 + image_mode: single + stitch_grid: + - 2 + - 2 + obs_image_size: + - 224 + - 224 + video_backend: torchvision_av + observation_indices: + - -4 + - -3 + - -2 + - -1 + - 0 + language_indices: + - 0 + state_indices: + - 0 + action_indices: + - 0 + - 1 + - 2 + - 3 + - 4 + - 5 + - 6 + - 7 + pack_image_sequence: true + image_sequence_length: 5 + context_images_column: observation.context_images +dataset: + source_hf: '' + config_name: null + source_subdir: null + converted_name: flappy_train__bridge + single_source_hf: '' + mixed_source_hf: '' + single_converted_name: flappy_train__bridge + mixed_converted_name: flappy_mixed_latency_train + single_latency_filter: null + mixed_latency_filter: null + force_download: false + setup_force: false + skip_verification: false + verify_rows: 200 + max_episodes: null + episodes_per_latency: null + latency_filter: null + debug_subset: + enabled: false + max_episodes: 5 + suffix: debug +base_model: + repo_id: Wan-AI/Wan2.2-TI2V-5B-Diffusers +initialization: + checkpoint_local_dir: playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1 + checkpoint_hf_repo_id: StarVLA/WM4A-Wan2d2-OFT-LIBERO-4in1 + checkpoint_filename: checkpoints/steps_60000_pytorch_model.pt +trainer: + max_train_steps: 2000 + num_warmup_steps: 0 + save_interval: 0 + eval_interval: 500 + eval_num_batches: 50 + per_latency_eval_num_batches: null + eval_action_classification: false + eval_action_classification_interval: null + cc_f1_tolerance: 1 + learning_rate: + base: 2.5e-05 + qwen_vl_interface: 1.0e-05 + action_model: 0.0001 + action_query_proj: 0.0001 + lr_scheduler_type: cosine_with_min_lr + scheduler_specific_kwargs: + min_lr: 1.0e-06 + freeze_modules: '' + freeze_vit: false + freeze_tied_embedding: false + freeze_llm_layers: [] + loss_scale: + vla: 1.0 + vlm: 0.1 + max_grad_norm: 1.0 + weight_decay: 0.0 + logging_frequency: 1 + profile_timing: + enabled: false + log_interval: 10 + gradient_clipping: 1.0 + gradient_accumulation_steps: 32 + distributed_backend: none + is_resume: false + pretrained_checkpoint: null + resume_step: 0 + reload_modules: null + optimizer: + name: AdamW + betas: + - 0.9 + - 0.95 + eps: 1.0e-08 + weight_decay: 1.0e-08 + fused: true + save_format: pt +workspace_dir: WORKSPACE_DIR +run_root_dir: /workspace/starVLA/results/Checkpoints +seed: 42 +wandb_entity: dongqianyu99-zhejiang-university +wandb_project: starVLA_rl_games +auth: + env_file: null + hf_token_env: HF_TOKEN + wandb_api_key_env: WANDB_API_KEY +paths: + run_root_dir: results/Checkpoints + dataset_local_dir: data/flappy_fix_latency_1_200ep_context5 + dataset_cache_dir: null + base_model_dir: playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + accelerate_config: starVLA/config/deepseeds/deepspeed_zero2.yaml +launch: + use_accelerate: true + gpus: null + num_processes: 1 + dry_run: false +conda: + enabled: true + env_name: null +rl_games: + model_alias: wan_oft + env_eval: + image_size: 224 + frameskip: 1 + seed: 42 + fixed_episode_seeds: true + latency_seed_stride: 0 + task_seed_stride: 0 + task_description: '' + action_chunk_execution: + enabled: false + chunk_size: null + enabled: true + distributed_mode: none + vectorized: + enabled: false + batch_size: 1 + latency: + prompt_map_path: data/flappy_fix_latency_1_200ep_context5/flappy_train__bridge/latency_prompt_map_eval_0_4.json + mode: single + values: + - 0 + mid_train: + enabled: true + interval_steps: 500 + latencies: + - 1 + num_episodes: 5 + max_steps_per_episode: 3600 + post_train: + enabled: true + latencies: + - 0 + - 1 + - 2 + - 3 + - 4 + num_episodes: 20 + max_steps_per_episode: 3600 + task: flappy + initialization_mode: bridge + action_carrier: bridge +model: wan_oft +env: flappy +init: wan_oft_libero +mode: single +checkpoint: + load: none + hf_repo_id: null + save_best_model: false + save_final_model: true + save_pt_file: true + save_training_state: false + local: + keep_last_n: 1 + sync: + enabled: false + repo_id: null + keep_last_n: 0 + sync_every_n_checkpoints: 1 + resume_policy: local_latest +run_id: wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce +output_dir: /workspace/starVLA/results/Checkpoints/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce +config_yaml: null +is_debug: false +version_id: '0.21' diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.yaml b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.yaml new file mode 100644 index 0000000000000000000000000000000000000000..e439205746d42aebd4223cf3e531ac6b0f211f99 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.yaml @@ -0,0 +1,113 @@ +checkpoint: + local: + keep_last_n: 1 + save_best_model: false + save_final_model: true + save_pt_file: true + save_training_state: false + sync: + enabled: false + keep_last_n: 0 + repo_id: null +datasets: + vla_data: + data_mix: flappy_train__bridge + dataset_py: lerobot_datasets + eval_data_mix: flappy_train__bridge__val + image_sequence_length: 5 + include_state: true + latency_curriculum: + enabled: false + obs_image_size: + - 224 + - 224 + pack_image_sequence: true + per_device_batch_size: 4 +framework: + action_model: + action_dim: 7 + action_env_dim: 2 + action_hidden_dim: 3072 + action_horizon: 8 + action_model_type: MLP + action_query_source: mean + class_weights: null + loss_type: current_discrete_ce + state_dim: 7 + name: WanOFT + qwenvl: + base_vlm: /workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + world_model: + base_wm: /workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + extract_layers: + - -1 + num_frames: 5 +output_dir: /workspace/starVLA/results/Checkpoints/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce +rl_games: + env_eval: + action_chunk_execution: + chunk_size: null + enabled: false + enabled: true + fixed_episode_seeds: true + frameskip: 1 + image_size: 224 + latency: + prompt_map_path: data/flappy_fix_latency_1_200ep_context5/flappy_train__bridge/latency_prompt_map_eval_0_4.json + latency_seed_stride: 0 + mid_train: + enabled: true + interval_steps: 500 + latencies: + - 1 + max_steps_per_episode: 3600 + num_episodes: 5 + seed: 42 + task_description: '' + task_seed_stride: 0 + vectorized: + enabled: false + model_alias: wan_oft + task: flappy +run_id: wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce +run_root_dir: /workspace/starVLA/results/Checkpoints +seed: 42 +trainer: + distributed_backend: none + eval_action_classification: false + eval_action_classification_interval: null + eval_interval: 500 + eval_num_batches: 50 + freeze_llm_layers: [] + freeze_modules: '' + freeze_tied_embedding: false + freeze_vit: false + gradient_accumulation_steps: 32 + gradient_clipping: 1.0 + is_resume: false + learning_rate: + action_model: 0.0001 + action_query_proj: 0.0001 + base: 2.5e-05 + qwen_vl_interface: 1.0e-05 + logging_frequency: 1 + lr_scheduler_type: cosine_with_min_lr + max_train_steps: 2000 + num_warmup_steps: 0 + optimizer: + betas: + - 0.9 + - 0.95 + eps: 1.0e-08 + fused: true + weight_decay: 1.0e-08 + per_latency_eval_num_batches: null + pretrained_checkpoint: null + profile_timing: + enabled: false + reload_modules: null + save_interval: 0 + scheduler_specific_kwargs: + min_lr: 1.0e-06 +wandb_entity: dongqianyu99-zhejiang-university +wandb_project: starVLA_rl_games diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/dataset_statistics.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/dataset_statistics.json new file mode 100644 index 0000000000000000000000000000000000000000..bdf8b3324aa996a468133627e1c3690eff3690ce --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/dataset_statistics.json @@ -0,0 +1,127 @@ +{ + "new_embodiment": { + "action": { + "mean": [ + 0.6979423761367798, + 0.3020575940608978, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "std": [ + 0.4588429629802704, + 0.4588429629802704, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "max": [ + 1.0, + 1.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "min": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q01": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q99": [ + 1.0, + 1.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "mask": [ + true, + true, + true, + true, + true, + true, + true + ] + }, + "state": { + "mean": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "std": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "max": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "min": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q01": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q99": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ] + }, + "num_transitions": 297483, + "num_trajectories": 180 + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/dataset_statistics_eval.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/dataset_statistics_eval.json new file mode 100644 index 0000000000000000000000000000000000000000..88b7fb21af76becf6ed0ffdc38ba13902347d226 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/dataset_statistics_eval.json @@ -0,0 +1,127 @@ +{ + "new_embodiment": { + "action": { + "mean": [ + 0.8606944680213928, + 0.13930556178092957, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "std": [ + 0.3463059663772583, + 0.3463059663772583, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "max": [ + 1.0, + 1.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "min": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q01": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q99": [ + 1.0, + 1.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "mask": [ + true, + true, + true, + true, + true, + true, + true + ] + }, + "state": { + "mean": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "std": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "max": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "min": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q01": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ], + "q99": [ + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0, + 0.0 + ] + }, + "num_transitions": 72000, + "num_trajectories": 20 + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_1000.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_1000.json new file mode 100644 index 0000000000000000000000000000000000000000..b2929500b085b72db26cfb5a1d69a99cdbd11257 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_1000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 15569, + "1": 2431 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_1500.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_1500.json new file mode 100644 index 0000000000000000000000000000000000000000..6a400c73727289187267220bee7c9e522f07533c --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_1500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 15535, + "1": 2465 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 1500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_2000.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..566e2cd16ed5129e2bfc35f61d1f19a06662f66b --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_2000.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 15564, + "1": 2436 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "task_count": 1, + "macro_mean_reward": 444.60000000002566, + "macro_mean_length": 3600.0, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_500.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_500.json new file mode 100644 index 0000000000000000000000000000000000000000..eb5ba877b025686efa4a661d98748a480c1ec333 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/mid_train/step_500.json @@ -0,0 +1,63 @@ +{ + "per_latency": { + "flappy/latency_1": { + "latency": 1, + "num_episodes": 5, + "mean_reward": 382.18000000001996, + "mean_length": 3096.8, + "std_reward": 124.8400000000113, + "std_length": 1006.4000000000002, + "episode_rewards": [ + 132.49999999999739, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 1084, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 13331, + "1": 2153 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4 + ] + } + }, + "aggregate": { + "stage": "mid_train", + "step": 500, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 5, + "mean_reward": 382.18000000001996, + "mean_length": 3096.8, + "std_reward": 124.8400000000113, + "std_length": 1006.4000000000002, + "task_count": 1, + "macro_mean_reward": 382.18000000001996, + "macro_mean_length": 3096.8, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/post_train/step_2000.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/post_train/step_2000.json new file mode 100644 index 0000000000000000000000000000000000000000..6e50c34f3a90be8e20569b3e36b38ad3a94f4931 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/eval/post_train/step_2000.json @@ -0,0 +1,531 @@ +{ + "per_latency": { + "flappy/latency_0": { + "latency": 0, + "num_episodes": 20, + "mean_reward": 72.23999999999896, + "mean_length": 597.95, + "std_reward": 63.11439930792145, + "std_length": 509.34550896223675, + "episode_rewards": [ + 57.800000000000445, + 34.80000000000019, + 20.800000000000036, + 174.7999999999955, + 81.29999999999968, + 169.79999999999572, + 132.79999999999737, + 11.29999999999998, + 34.80000000000019, + 6.799999999999992, + 53.300000000000395, + 34.80000000000019, + 76.79999999999988, + 71.80000000000011, + 20.700000000000035, + 20.800000000000036, + 183.7999999999951, + 207.29999999999404, + 20.700000000000035, + 29.800000000000136 + ], + "episode_lengths": [ + 481, + 296, + 183, + 1426, + 671, + 1385, + 1087, + 106, + 296, + 70, + 445, + 296, + 635, + 594, + 182, + 183, + 1498, + 1688, + 182, + 255 + ], + "decoded_action_hist": { + "1": 903, + "0": 11056 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_1": { + "latency": 1, + "num_episodes": 20, + "mean_reward": 444.60000000002566, + "mean_length": 3600.0, + "std_reward": 0.0, + "std_length": 0.0, + "episode_rewards": [ + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566, + 444.60000000002566 + ], + "episode_lengths": [ + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600, + 3600 + ], + "decoded_action_hist": { + "0": 62110, + "1": 9890 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_2": { + "latency": 2, + "num_episodes": 20, + "mean_reward": 6.615000000000006, + "mean_length": 71.75, + "std_reward": 7.366700414704039, + "std_length": 59.72342505248673, + "episode_rewards": [ + 3.8999999999999986, + 4.099999999999998, + 4.1999999999999975, + 37.20000000000022, + 4.099999999999998, + 3.8999999999999986, + 5.999999999999995, + 3.8999999999999986, + 4.299999999999997, + 3.9999999999999982, + 13.299999999999972, + 4.299999999999997, + 3.8999999999999986, + 8.999999999999984, + 3.8999999999999986, + 3.8999999999999986, + 4.099999999999998, + 4.299999999999997, + 5.999999999999995, + 3.9999999999999982 + ], + "episode_lengths": [ + 50, + 52, + 53, + 320, + 52, + 50, + 62, + 50, + 54, + 51, + 126, + 54, + 50, + 92, + 50, + 50, + 52, + 54, + 62, + 51 + ], + "decoded_action_hist": { + "0": 1186, + "1": 249 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_3": { + "latency": 3, + "num_episodes": 20, + "mean_reward": 4.689999999999996, + "mean_length": 56.55, + "std_reward": 0.6579513659838381, + "std_length": 3.9430318284284747, + "episode_rewards": [ + 3.9999999999999982, + 4.699999999999996, + 4.799999999999995, + 4.799999999999995, + 4.4999999999999964, + 3.9999999999999982, + 6.099999999999994, + 3.9999999999999982, + 4.099999999999998, + 4.399999999999997, + 4.799999999999995, + 4.699999999999996, + 4.299999999999997, + 4.799999999999995, + 6.099999999999994, + 3.9999999999999982, + 4.4999999999999964, + 4.699999999999996, + 6.099999999999994, + 4.399999999999997 + ], + "episode_lengths": [ + 51, + 58, + 59, + 59, + 56, + 51, + 63, + 51, + 52, + 55, + 59, + 58, + 54, + 59, + 63, + 51, + 56, + 58, + 63, + 55 + ], + "decoded_action_hist": { + "0": 934, + "1": 197 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + }, + "flappy/latency_4": { + "latency": 4, + "num_episodes": 20, + "mean_reward": 5.044999999999996, + "mean_length": 60.55, + "std_reward": 1.1386285610329618, + "std_length": 8.749142815156237, + "episode_rewards": [ + 4.799999999999995, + 4.799999999999995, + 4.399999999999997, + 4.599999999999996, + 4.799999999999995, + 4.999999999999995, + 8.399999999999986, + 4.999999999999995, + 4.299999999999997, + 4.699999999999996, + 4.599999999999996, + 4.299999999999997, + 4.599999999999996, + 4.599999999999996, + 4.799999999999995, + 4.999999999999995, + 4.799999999999995, + 4.299999999999997, + 8.399999999999986, + 4.699999999999996 + ], + "episode_lengths": [ + 59, + 59, + 55, + 57, + 59, + 61, + 86, + 61, + 54, + 58, + 57, + 54, + 57, + 57, + 59, + 61, + 59, + 54, + 86, + 58 + ], + "decoded_action_hist": { + "0": 960, + "1": 251 + }, + "fixed_episode_seeds": true, + "eval_seed": 42, + "episode_seeds": [ + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61 + ], + "episode_indices": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19 + ] + } + }, + "aggregate": { + "stage": "post_train", + "step": 2000, + "task": "flappy", + "model_alias": "wan_oft", + "fixed_episode_seeds": true, + "eval_seed": 42, + "total_episodes": 100, + "mean_reward": 106.63800000000492, + "mean_length": 877.36, + "std_reward": 173.2974862945318, + "std_length": 1395.9831984662278, + "task_count": 1, + "macro_mean_reward": 106.6380000000049, + "macro_mean_length": 877.36, + "distributed_eval": false + } +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/config.yaml b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/config.yaml new file mode 100644 index 0000000000000000000000000000000000000000..8f1f3238128f806683533bffb66bd7199afb833d --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/config.yaml @@ -0,0 +1,258 @@ +framework: + qwenvl: + base_vlm: /workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + attn_implementation: sdpa + enable_gradient_checkpointing: true + action_model: + state_dim: 7 + loss_type: current_discrete_ce + action_horizon: 8 + future_action_window_size: 7 + past_action_window_size: 0 + action_model_type: MLP + action_dim: 7 + action_env_dim: 2 + action_query_source: mean + action_query_num_heads: 24 + name: WanOFT + world_model: + base_wm: /workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + extract_layers: + - -1 + num_frames: 5 +datasets: + vla_data: + dataset_py: lerobot_datasets + include_state: true + data_root_dir: /workspace/starVLA/data/flappy_fix_latency_1_200ep_context5 + data_mix: flappy_train__bridge + eval_data_mix: flappy_train__bridge__val + custom_mixtures_path: null + action_type: discrete + sequential_step_sampling: false + eval_sequential_step_sampling: null + num_workers: 8 + eval_num_workers: 8 + prefetch_factor: 4 + persistent_workers: true + pin_memory: true + shuffle: true + action_balance: + enabled: false + strategy: balanced_epoch + action_key: action_id + target_flap_fraction: 0.3 + noop_id: 0 + flap_id: 1 + latency_curriculum: + enabled: false + strategy: exclusive + latencies: null + phase_steps: null + new_latency_passes: 1.0 + replay_passes: 0.25 + target_total_passes: 2.0 + final_equalization: true + step_budget_mode: auto + eval_at_phase_end: false + save_at_phase_end: false + computed_plan: null + effective_batch_size: null + per_device_batch_size: 4 + load_all_data_for_training: true + num_obs_frames: 1 + image_mode: single + stitch_grid: + - 2 + - 2 + obs_image_size: + - 224 + - 224 + video_backend: torchvision_av + observation_indices: + - -4 + - -3 + - -2 + - -1 + - 0 + language_indices: + - 0 + state_indices: + - 0 + action_indices: + - 0 + - 1 + - 2 + - 3 + - 4 + - 5 + - 6 + - 7 + pack_image_sequence: true + image_sequence_length: 5 + context_images_column: observation.context_images +dataset: + source_hf: '' + config_name: null + source_subdir: null + converted_name: flappy_train__bridge + single_source_hf: '' + mixed_source_hf: '' + single_converted_name: flappy_train__bridge + mixed_converted_name: flappy_mixed_latency_train + single_latency_filter: null + mixed_latency_filter: null + force_download: false + setup_force: false + skip_verification: false + verify_rows: 200 + max_episodes: null + episodes_per_latency: null + latency_filter: null + debug_subset: + enabled: false + max_episodes: 5 + suffix: debug +base_model: + repo_id: Wan-AI/Wan2.2-TI2V-5B-Diffusers +initialization: + checkpoint_local_dir: playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1 + checkpoint_hf_repo_id: StarVLA/WM4A-Wan2d2-OFT-LIBERO-4in1 + checkpoint_filename: checkpoints/steps_60000_pytorch_model.pt +trainer: + max_train_steps: 2000 + num_warmup_steps: 0 + save_interval: 0 + eval_interval: 500 + eval_num_batches: 50 + per_latency_eval_num_batches: null + eval_action_classification: false + eval_action_classification_interval: null + cc_f1_tolerance: 1 + learning_rate: + base: 2.5e-05 + qwen_vl_interface: 1.0e-05 + action_model: 0.0001 + action_query_proj: 0.0001 + lr_scheduler_type: cosine_with_min_lr + scheduler_specific_kwargs: + min_lr: 1.0e-06 + freeze_modules: '' + freeze_vit: false + freeze_tied_embedding: false + freeze_llm_layers: [] + loss_scale: + vla: 1.0 + vlm: 0.1 + max_grad_norm: 1.0 + weight_decay: 0.0 + logging_frequency: 1 + profile_timing: + enabled: false + log_interval: 10 + gradient_clipping: 1.0 + gradient_accumulation_steps: 32 + distributed_backend: none + is_resume: false + pretrained_checkpoint: /workspace/starVLA/playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1/checkpoints/steps_60000_pytorch_model.pt + resume_step: 0 + reload_modules: null + optimizer: + name: AdamW + betas: + - 0.9 + - 0.95 + eps: 1.0e-08 + weight_decay: 1.0e-08 + fused: true + save_format: pt +workspace_dir: WORKSPACE_DIR +run_root_dir: /workspace/starVLA/results/Checkpoints +seed: 42 +wandb_entity: dongqianyu99-zhejiang-university +wandb_project: starVLA_rl_games +auth: + env_file: null + hf_token_env: HF_TOKEN + wandb_api_key_env: WANDB_API_KEY +paths: + run_root_dir: results/Checkpoints + dataset_local_dir: data/flappy_fix_latency_1_200ep_context5 + dataset_cache_dir: null + base_model_dir: playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + accelerate_config: starVLA/config/deepseeds/deepspeed_zero2.yaml +launch: + use_accelerate: true + gpus: null + num_processes: 1 + dry_run: false +conda: + enabled: true + env_name: null +rl_games: + model_alias: wan_oft + env_eval: + image_size: 224 + frameskip: 1 + seed: 42 + fixed_episode_seeds: true + latency_seed_stride: 0 + task_seed_stride: 0 + task_description: '' + action_chunk_execution: + enabled: false + chunk_size: null + enabled: true + distributed_mode: none + vectorized: + enabled: false + batch_size: 1 + latency: + prompt_map_path: data/flappy_fix_latency_1_200ep_context5/flappy_train__bridge/latency_prompt_map_eval_0_4.json + mode: single + values: + - 0 + mid_train: + enabled: true + interval_steps: 500 + latencies: + - 1 + num_episodes: 5 + max_steps_per_episode: 3600 + post_train: + enabled: true + latencies: + - 0 + - 1 + - 2 + - 3 + - 4 + num_episodes: 20 + max_steps_per_episode: 3600 + task: flappy + initialization_mode: bridge + action_carrier: bridge +model: wan_oft +env: flappy +init: wan_oft_libero +mode: single +checkpoint: + load: none + hf_repo_id: null + save_best_model: false + save_final_model: true + save_pt_file: true + save_training_state: false + local: + keep_last_n: 1 + sync: + enabled: false + repo_id: null + keep_last_n: 0 + sync_every_n_checkpoints: 1 + resume_policy: local_latest +run_id: wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce +output_dir: null +config_yaml: null +is_debug: false +version_id: 0.21 diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/hydra.yaml b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/hydra.yaml new file mode 100644 index 0000000000000000000000000000000000000000..a9406448c788cba8212132b034b38ecf79cffdc9 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/hydra.yaml @@ -0,0 +1,356 @@ +hydra: + run: + dir: ${run_root_dir}/${run_id}/hydra + sweep: + dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S} + subdir: ${hydra.job.num} + launcher: + _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher + sweeper: + _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper + max_batch_size: null + params: null + help: + app_name: ${hydra.job.name} + header: '${hydra.help.app_name} is powered by Hydra. + + ' + footer: 'Powered by Hydra (https://hydra.cc) + + Use --hydra-help to view Hydra specific help + + ' + template: '${hydra.help.header} + + == Configuration groups == + + Compose your configuration from those groups (group=option) + + + $APP_CONFIG_GROUPS + + + == Config == + + Override anything in the config (foo.bar=value) + + + $CONFIG + + + ${hydra.help.footer} + + ' + hydra_help: + template: 'Hydra (${hydra.runtime.version}) + + See https://hydra.cc for more info. + + + == Flags == + + $FLAGS_HELP + + + == Configuration groups == + + Compose your configuration from those groups (For example, append hydra/job_logging=disabled + to command line) + + + $HYDRA_CONFIG_GROUPS + + + Use ''--cfg hydra'' to Show the Hydra config. + + ' + hydra_help: ??? + hydra_logging: + version: 1 + formatters: + simple: + format: '[%(asctime)s][HYDRA] %(message)s' + handlers: + console: + class: logging.StreamHandler + formatter: simple + stream: ext://sys.stdout + root: + level: INFO + handlers: + - console + loggers: + logging_example: + level: DEBUG + disable_existing_loggers: false + job_logging: + version: 1 + formatters: + simple: + format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s' + handlers: + console: + class: logging.StreamHandler + formatter: simple + stream: ext://sys.stdout + file: + class: logging.FileHandler + formatter: simple + filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log + root: + level: INFO + handlers: + - console + - file + disable_existing_loggers: false + env: {} + mode: RUN + searchpath: [] + callbacks: {} + output_subdir: .hydra + overrides: + hydra: + - hydra.mode=RUN + task: + - model=wan_oft + - env=flappy + - init=wan_oft_libero + - mode=single + - ++framework.qwenvl.base_vlm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + - ++framework.qwenvl.attn_implementation=sdpa + - ++framework.qwenvl.enable_gradient_checkpointing=true + - ++framework.action_model.state_dim=7 + - ++framework.action_model.loss_type=current_discrete_ce + - ++framework.action_model.action_horizon=8 + - ++framework.action_model.future_action_window_size=7 + - ++framework.action_model.past_action_window_size=0 + - ++framework.action_model.action_model_type=MLP + - ++framework.action_model.action_dim=7 + - ++framework.action_model.action_env_dim=2 + - ++framework.action_model.action_query_source=mean + - ++framework.action_model.action_query_num_heads=24 + - ++framework.name=WanOFT + - ++framework.world_model.base_wm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + - ++framework.world_model.extract_layers=[-1] + - ++framework.world_model.num_frames=5 + - ++datasets.vla_data.dataset_py=lerobot_datasets + - ++datasets.vla_data.include_state=true + - ++datasets.vla_data.data_root_dir=playground/Datasets/rl_games + - ++datasets.vla_data.data_mix=flappy_train__bridge + - ++datasets.vla_data.eval_data_mix=flappy_train__bridge__val + - ++datasets.vla_data.custom_mixtures_path=null + - ++datasets.vla_data.action_type=discrete + - ++datasets.vla_data.sequential_step_sampling=false + - ++datasets.vla_data.eval_sequential_step_sampling=null + - ++datasets.vla_data.num_workers=8 + - ++datasets.vla_data.eval_num_workers=8 + - ++datasets.vla_data.prefetch_factor=4 + - ++datasets.vla_data.persistent_workers=true + - ++datasets.vla_data.pin_memory=true + - ++datasets.vla_data.shuffle=true + - ++datasets.vla_data.action_balance.enabled=false + - ++datasets.vla_data.action_balance.strategy=balanced_epoch + - ++datasets.vla_data.action_balance.action_key=action_id + - ++datasets.vla_data.action_balance.target_flap_fraction=0.3 + - ++datasets.vla_data.action_balance.noop_id=0 + - ++datasets.vla_data.action_balance.flap_id=1 + - ++datasets.vla_data.latency_curriculum.enabled=false + - ++datasets.vla_data.latency_curriculum.strategy=exclusive + - ++datasets.vla_data.latency_curriculum.latencies=null + - ++datasets.vla_data.latency_curriculum.phase_steps=null + - ++datasets.vla_data.latency_curriculum.new_latency_passes=1.0 + - ++datasets.vla_data.latency_curriculum.replay_passes=0.25 + - ++datasets.vla_data.latency_curriculum.target_total_passes=2.0 + - ++datasets.vla_data.latency_curriculum.final_equalization=true + - ++datasets.vla_data.latency_curriculum.step_budget_mode=auto + - ++datasets.vla_data.latency_curriculum.eval_at_phase_end=false + - ++datasets.vla_data.latency_curriculum.save_at_phase_end=false + - ++datasets.vla_data.latency_curriculum.computed_plan=null + - ++datasets.vla_data.latency_curriculum.effective_batch_size=null + - ++datasets.vla_data.per_device_batch_size=4 + - ++datasets.vla_data.load_all_data_for_training=true + - ++datasets.vla_data.num_obs_frames=1 + - ++datasets.vla_data.image_mode=single + - ++datasets.vla_data.stitch_grid=[2,2] + - ++datasets.vla_data.obs_image_size=[224,224] + - ++datasets.vla_data.video_backend=torchvision_av + - ++datasets.vla_data.observation_indices=[-4,-3,-2,-1,0] + - ++datasets.vla_data.language_indices=[0] + - ++datasets.vla_data.state_indices=[0] + - ++datasets.vla_data.action_indices=[0,1,2,3,4,5,6,7] + - ++datasets.vla_data.pack_image_sequence=true + - ++datasets.vla_data.image_sequence_length=5 + - ++datasets.vla_data.context_images_column=observation.context_images + - ++dataset.source_hf= + - ++dataset.config_name=null + - ++dataset.source_subdir=null + - ++dataset.converted_name=flappy_train__bridge + - ++dataset.single_source_hf= + - ++dataset.mixed_source_hf= + - ++dataset.single_converted_name=flappy_train__bridge + - ++dataset.mixed_converted_name=flappy_mixed_latency_train + - ++dataset.single_latency_filter=null + - ++dataset.mixed_latency_filter=null + - ++dataset.force_download=false + - ++dataset.setup_force=false + - ++dataset.skip_verification=false + - ++dataset.verify_rows=200 + - ++dataset.max_episodes=null + - ++dataset.episodes_per_latency=null + - ++dataset.latency_filter=null + - ++dataset.debug_subset.enabled=false + - ++dataset.debug_subset.max_episodes=5 + - ++dataset.debug_subset.suffix=debug + - ++base_model.repo_id=Wan-AI/Wan2.2-TI2V-5B-Diffusers + - ++initialization.checkpoint_local_dir=playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1 + - ++initialization.checkpoint_hf_repo_id=StarVLA/WM4A-Wan2d2-OFT-LIBERO-4in1 + - ++initialization.checkpoint_filename=checkpoints/steps_60000_pytorch_model.pt + - ++trainer.max_train_steps=2000 + - ++trainer.num_warmup_steps=0 + - ++trainer.save_interval=0 + - ++trainer.eval_interval=500 + - ++trainer.eval_num_batches=50 + - ++trainer.per_latency_eval_num_batches=null + - ++trainer.eval_action_classification=false + - ++trainer.eval_action_classification_interval=null + - ++trainer.cc_f1_tolerance=1 + - ++trainer.learning_rate.base=2.5e-05 + - ++trainer.learning_rate.qwen_vl_interface=1e-05 + - ++trainer.learning_rate.action_model=0.0001 + - ++trainer.learning_rate.action_query_proj=0.0001 + - ++trainer.lr_scheduler_type=cosine_with_min_lr + - ++trainer.scheduler_specific_kwargs.min_lr=1e-06 + - ++trainer.freeze_modules= + - ++trainer.freeze_vit=false + - ++trainer.freeze_tied_embedding=false + - ++trainer.freeze_llm_layers=[] + - ++trainer.loss_scale.vla=1.0 + - ++trainer.loss_scale.vlm=0.1 + - ++trainer.max_grad_norm=1.0 + - ++trainer.weight_decay=0.0 + - ++trainer.logging_frequency=1 + - ++trainer.profile_timing.enabled=false + - ++trainer.profile_timing.log_interval=10 + - ++trainer.gradient_clipping=1.0 + - ++trainer.gradient_accumulation_steps=32 + - ++trainer.distributed_backend=none + - ++trainer.is_resume=false + - ++trainer.pretrained_checkpoint=null + - ++trainer.resume_step=0 + - ++trainer.reload_modules=null + - ++trainer.optimizer.name=AdamW + - ++trainer.optimizer.betas=[0.9,0.95] + - ++trainer.optimizer.eps=1e-08 + - ++trainer.optimizer.weight_decay=1e-08 + - ++trainer.optimizer.fused=true + - ++trainer.save_format=pt + - ++workspace_dir=WORKSPACE_DIR + - ++run_root_dir=results/Checkpoints + - ++seed=42 + - ++auth.env_file=null + - ++auth.hf_token_env=HF_TOKEN + - ++auth.wandb_api_key_env=WANDB_API_KEY + - ++paths.run_root_dir=results/Checkpoints + - ++paths.dataset_local_dir=data/flappy_fix_latency_1_200ep_context5 + - ++paths.dataset_cache_dir=null + - ++paths.base_model_dir=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + - ++paths.accelerate_config=starVLA/config/deepseeds/deepspeed_zero2.yaml + - ++rl_games.model_alias=wan_oft + - ++rl_games.env_eval.image_size=224 + - ++rl_games.env_eval.frameskip=1 + - ++rl_games.env_eval.seed=42 + - ++rl_games.env_eval.fixed_episode_seeds=true + - ++rl_games.env_eval.latency_seed_stride=0 + - ++rl_games.env_eval.task_seed_stride=0 + - ++rl_games.env_eval.task_description= + - ++rl_games.env_eval.action_chunk_execution.enabled=false + - ++rl_games.env_eval.action_chunk_execution.chunk_size=null + - ++rl_games.env_eval.enabled=true + - ++rl_games.env_eval.distributed_mode=none + - ++rl_games.env_eval.vectorized.enabled=false + - ++rl_games.env_eval.vectorized.batch_size=1 + - ++rl_games.env_eval.latency.prompt_map_path=data/flappy_fix_latency_1_200ep_context5/flappy_train__bridge/latency_prompt_map_eval_0_4.json + - ++rl_games.env_eval.latency.mode=single + - ++rl_games.env_eval.latency.values=[0] + - ++rl_games.env_eval.mid_train.enabled=true + - ++rl_games.env_eval.mid_train.interval_steps=500 + - ++rl_games.env_eval.mid_train.latencies=[1] + - ++rl_games.env_eval.mid_train.num_episodes=5 + - ++rl_games.env_eval.mid_train.max_steps_per_episode=3600 + - ++rl_games.env_eval.post_train.enabled=true + - ++rl_games.env_eval.post_train.latencies=[0,1,2,3,4] + - ++rl_games.env_eval.post_train.num_episodes=20 + - ++rl_games.env_eval.post_train.max_steps_per_episode=3600 + - ++rl_games.task=flappy + - ++rl_games.initialization_mode=bridge + - ++rl_games.action_carrier=bridge + - ++checkpoint.load=none + - ++checkpoint.hf_repo_id=null + - ++checkpoint.save_best_model=false + - ++checkpoint.save_final_model=true + - ++checkpoint.save_pt_file=true + - ++checkpoint.save_training_state=false + - ++checkpoint.local.keep_last_n=1 + - ++checkpoint.sync.enabled=false + - ++checkpoint.sync.repo_id=null + - ++checkpoint.sync.keep_last_n=0 + - ++checkpoint.sync.sync_every_n_checkpoints=1 + - ++checkpoint.sync.resume_policy=local_latest + - ++run_id=wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce + - ++output_dir=null + - ++config_yaml=null + - ++is_debug=false + - ++version_id=0.21 + - ++run_root_dir=/workspace/starVLA/results/Checkpoints + - ++trainer.is_resume=false + - ++trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1/checkpoints/steps_60000_pytorch_model.pt + - ++trainer.resume_step=0 + - ++datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1_200ep_context5 + - ++framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + - ++framework.world_model.base_wm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers + job: + name: train_starvla_hydra + chdir: false + override_dirname: ++auth.env_file=null,++auth.hf_token_env=HF_TOKEN,++auth.wandb_api_key_env=WANDB_API_KEY,++base_model.repo_id=Wan-AI/Wan2.2-TI2V-5B-Diffusers,++checkpoint.hf_repo_id=null,++checkpoint.load=none,++checkpoint.local.keep_last_n=1,++checkpoint.save_best_model=false,++checkpoint.save_final_model=true,++checkpoint.save_pt_file=true,++checkpoint.save_training_state=false,++checkpoint.sync.enabled=false,++checkpoint.sync.keep_last_n=0,++checkpoint.sync.repo_id=null,++checkpoint.sync.resume_policy=local_latest,++checkpoint.sync.sync_every_n_checkpoints=1,++config_yaml=null,++dataset.config_name=null,++dataset.converted_name=flappy_train__bridge,++dataset.debug_subset.enabled=false,++dataset.debug_subset.max_episodes=5,++dataset.debug_subset.suffix=debug,++dataset.episodes_per_latency=null,++dataset.force_download=false,++dataset.latency_filter=null,++dataset.max_episodes=null,++dataset.mixed_converted_name=flappy_mixed_latency_train,++dataset.mixed_latency_filter=null,++dataset.mixed_source_hf=,++dataset.setup_force=false,++dataset.single_converted_name=flappy_train__bridge,++dataset.single_latency_filter=null,++dataset.single_source_hf=,++dataset.skip_verification=false,++dataset.source_hf=,++dataset.source_subdir=null,++dataset.verify_rows=200,++datasets.vla_data.action_balance.action_key=action_id,++datasets.vla_data.action_balance.enabled=false,++datasets.vla_data.action_balance.flap_id=1,++datasets.vla_data.action_balance.noop_id=0,++datasets.vla_data.action_balance.strategy=balanced_epoch,++datasets.vla_data.action_balance.target_flap_fraction=0.3,++datasets.vla_data.action_indices=[0,1,2,3,4,5,6,7],++datasets.vla_data.action_type=discrete,++datasets.vla_data.context_images_column=observation.context_images,++datasets.vla_data.custom_mixtures_path=null,++datasets.vla_data.data_mix=flappy_train__bridge,++datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1_200ep_context5,++datasets.vla_data.data_root_dir=playground/Datasets/rl_games,++datasets.vla_data.dataset_py=lerobot_datasets,++datasets.vla_data.eval_data_mix=flappy_train__bridge__val,++datasets.vla_data.eval_num_workers=8,++datasets.vla_data.eval_sequential_step_sampling=null,++datasets.vla_data.image_mode=single,++datasets.vla_data.image_sequence_length=5,++datasets.vla_data.include_state=true,++datasets.vla_data.language_indices=[0],++datasets.vla_data.latency_curriculum.computed_plan=null,++datasets.vla_data.latency_curriculum.effective_batch_size=null,++datasets.vla_data.latency_curriculum.enabled=false,++datasets.vla_data.latency_curriculum.eval_at_phase_end=false,++datasets.vla_data.latency_curriculum.final_equalization=true,++datasets.vla_data.latency_curriculum.latencies=null,++datasets.vla_data.latency_curriculum.new_latency_passes=1.0,++datasets.vla_data.latency_curriculum.phase_steps=null,++datasets.vla_data.latency_curriculum.replay_passes=0.25,++datasets.vla_data.latency_curriculum.save_at_phase_end=false,++datasets.vla_data.latency_curriculum.step_budget_mode=auto,++datasets.vla_data.latency_curriculum.strategy=exclusive,++datasets.vla_data.latency_curriculum.target_total_passes=2.0,++datasets.vla_data.load_all_data_for_training=true,++datasets.vla_data.num_obs_frames=1,++datasets.vla_data.num_workers=8,++datasets.vla_data.obs_image_size=[224,224],++datasets.vla_data.observation_indices=[-4,-3,-2,-1,0],++datasets.vla_data.pack_image_sequence=true,++datasets.vla_data.per_device_batch_size=4,++datasets.vla_data.persistent_workers=true,++datasets.vla_data.pin_memory=true,++datasets.vla_data.prefetch_factor=4,++datasets.vla_data.sequential_step_sampling=false,++datasets.vla_data.shuffle=true,++datasets.vla_data.state_indices=[0],++datasets.vla_data.stitch_grid=[2,2],++datasets.vla_data.video_backend=torchvision_av,++framework.action_model.action_dim=7,++framework.action_model.action_env_dim=2,++framework.action_model.action_horizon=8,++framework.action_model.action_model_type=MLP,++framework.action_model.action_query_num_heads=24,++framework.action_model.action_query_source=mean,++framework.action_model.future_action_window_size=7,++framework.action_model.loss_type=current_discrete_ce,++framework.action_model.past_action_window_size=0,++framework.action_model.state_dim=7,++framework.name=WanOFT,++framework.qwenvl.attn_implementation=sdpa,++framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers,++framework.qwenvl.base_vlm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers,++framework.qwenvl.enable_gradient_checkpointing=true,++framework.world_model.base_wm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers,++framework.world_model.base_wm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers,++framework.world_model.extract_layers=[-1],++framework.world_model.num_frames=5,++initialization.checkpoint_filename=checkpoints/steps_60000_pytorch_model.pt,++initialization.checkpoint_hf_repo_id=StarVLA/WM4A-Wan2d2-OFT-LIBERO-4in1,++initialization.checkpoint_local_dir=playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1,++is_debug=false,++output_dir=null,++paths.accelerate_config=starVLA/config/deepseeds/deepspeed_zero2.yaml,++paths.base_model_dir=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers,++paths.dataset_cache_dir=null,++paths.dataset_local_dir=data/flappy_fix_latency_1_200ep_context5,++paths.run_root_dir=results/Checkpoints,++rl_games.action_carrier=bridge,++rl_games.env_eval.action_chunk_execution.chunk_size=null,++rl_games.env_eval.action_chunk_execution.enabled=false,++rl_games.env_eval.distributed_mode=none,++rl_games.env_eval.enabled=true,++rl_games.env_eval.fixed_episode_seeds=true,++rl_games.env_eval.frameskip=1,++rl_games.env_eval.image_size=224,++rl_games.env_eval.latency.mode=single,++rl_games.env_eval.latency.prompt_map_path=data/flappy_fix_latency_1_200ep_context5/flappy_train__bridge/latency_prompt_map_eval_0_4.json,++rl_games.env_eval.latency.values=[0],++rl_games.env_eval.latency_seed_stride=0,++rl_games.env_eval.mid_train.enabled=true,++rl_games.env_eval.mid_train.interval_steps=500,++rl_games.env_eval.mid_train.latencies=[1],++rl_games.env_eval.mid_train.max_steps_per_episode=3600,++rl_games.env_eval.mid_train.num_episodes=5,++rl_games.env_eval.post_train.enabled=true,++rl_games.env_eval.post_train.latencies=[0,1,2,3,4],++rl_games.env_eval.post_train.max_steps_per_episode=3600,++rl_games.env_eval.post_train.num_episodes=20,++rl_games.env_eval.seed=42,++rl_games.env_eval.task_description=,++rl_games.env_eval.task_seed_stride=0,++rl_games.env_eval.vectorized.batch_size=1,++rl_games.env_eval.vectorized.enabled=false,++rl_games.initialization_mode=bridge,++rl_games.model_alias=wan_oft,++rl_games.task=flappy,++run_id=wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce,++run_root_dir=/workspace/starVLA/results/Checkpoints,++run_root_dir=results/Checkpoints,++seed=42,++trainer.cc_f1_tolerance=1,++trainer.distributed_backend=none,++trainer.eval_action_classification=false,++trainer.eval_action_classification_interval=null,++trainer.eval_interval=500,++trainer.eval_num_batches=50,++trainer.freeze_llm_layers=[],++trainer.freeze_modules=,++trainer.freeze_tied_embedding=false,++trainer.freeze_vit=false,++trainer.gradient_accumulation_steps=32,++trainer.gradient_clipping=1.0,++trainer.is_resume=false,++trainer.is_resume=false,++trainer.learning_rate.action_model=0.0001,++trainer.learning_rate.action_query_proj=0.0001,++trainer.learning_rate.base=2.5e-05,++trainer.learning_rate.qwen_vl_interface=1e-05,++trainer.logging_frequency=1,++trainer.loss_scale.vla=1.0,++trainer.loss_scale.vlm=0.1,++trainer.lr_scheduler_type=cosine_with_min_lr,++trainer.max_grad_norm=1.0,++trainer.max_train_steps=2000,++trainer.num_warmup_steps=0,++trainer.optimizer.betas=[0.9,0.95],++trainer.optimizer.eps=1e-08,++trainer.optimizer.fused=true,++trainer.optimizer.name=AdamW,++trainer.optimizer.weight_decay=1e-08,++trainer.per_latency_eval_num_batches=null,++trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1/checkpoints/steps_60000_pytorch_model.pt,++trainer.pretrained_checkpoint=null,++trainer.profile_timing.enabled=false,++trainer.profile_timing.log_interval=10,++trainer.reload_modules=null,++trainer.resume_step=0,++trainer.resume_step=0,++trainer.save_format=pt,++trainer.save_interval=0,++trainer.scheduler_specific_kwargs.min_lr=1e-06,++trainer.weight_decay=0.0,++version_id=0.21,++workspace_dir=WORKSPACE_DIR,env=flappy,init=wan_oft_libero,mode=single,model=wan_oft + id: ??? + num: ??? + config_name: train + env_set: {} + env_copy: [] + config: + override_dirname: + kv_sep: '=' + item_sep: ',' + exclude_keys: [] + runtime: + version: 1.3.3 + version_base: '1.1' + cwd: /workspace/starVLA + config_sources: + - path: hydra.conf + schema: pkg + provider: hydra + - path: /workspace/starVLA/examples/rl_games/config + schema: file + provider: main + - path: '' + schema: structured + provider: schema + output_dir: /workspace/starVLA/results/Checkpoints/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra + choices: + cross_task_setup: null + checkpoint: default + mode: single + init: wan_oft_libero + env: flappy + model: wan_oft + hydra/env: default + hydra/callbacks: null + hydra/job_logging: default + hydra/hydra_logging: default + hydra/hydra_help: default + hydra/help: default + hydra/sweeper: basic + hydra/launcher: basic + hydra/output: default + verbose: false diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/overrides.yaml b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/overrides.yaml new file mode 100644 index 0000000000000000000000000000000000000000..fd4076888ce2fcd4bf13b00e888f537252f0467f --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/.hydra/overrides.yaml @@ -0,0 +1,196 @@ +- model=wan_oft +- env=flappy +- init=wan_oft_libero +- mode=single +- ++framework.qwenvl.base_vlm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers +- ++framework.qwenvl.attn_implementation=sdpa +- ++framework.qwenvl.enable_gradient_checkpointing=true +- ++framework.action_model.state_dim=7 +- ++framework.action_model.loss_type=current_discrete_ce +- ++framework.action_model.action_horizon=8 +- ++framework.action_model.future_action_window_size=7 +- ++framework.action_model.past_action_window_size=0 +- ++framework.action_model.action_model_type=MLP +- ++framework.action_model.action_dim=7 +- ++framework.action_model.action_env_dim=2 +- ++framework.action_model.action_query_source=mean +- ++framework.action_model.action_query_num_heads=24 +- ++framework.name=WanOFT +- ++framework.world_model.base_wm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers +- ++framework.world_model.extract_layers=[-1] +- ++framework.world_model.num_frames=5 +- ++datasets.vla_data.dataset_py=lerobot_datasets +- ++datasets.vla_data.include_state=true +- ++datasets.vla_data.data_root_dir=playground/Datasets/rl_games +- ++datasets.vla_data.data_mix=flappy_train__bridge +- ++datasets.vla_data.eval_data_mix=flappy_train__bridge__val +- ++datasets.vla_data.custom_mixtures_path=null +- ++datasets.vla_data.action_type=discrete +- ++datasets.vla_data.sequential_step_sampling=false +- ++datasets.vla_data.eval_sequential_step_sampling=null +- ++datasets.vla_data.num_workers=8 +- ++datasets.vla_data.eval_num_workers=8 +- ++datasets.vla_data.prefetch_factor=4 +- ++datasets.vla_data.persistent_workers=true +- ++datasets.vla_data.pin_memory=true +- ++datasets.vla_data.shuffle=true +- ++datasets.vla_data.action_balance.enabled=false +- ++datasets.vla_data.action_balance.strategy=balanced_epoch +- ++datasets.vla_data.action_balance.action_key=action_id +- ++datasets.vla_data.action_balance.target_flap_fraction=0.3 +- ++datasets.vla_data.action_balance.noop_id=0 +- ++datasets.vla_data.action_balance.flap_id=1 +- ++datasets.vla_data.latency_curriculum.enabled=false +- ++datasets.vla_data.latency_curriculum.strategy=exclusive +- ++datasets.vla_data.latency_curriculum.latencies=null +- ++datasets.vla_data.latency_curriculum.phase_steps=null +- ++datasets.vla_data.latency_curriculum.new_latency_passes=1.0 +- ++datasets.vla_data.latency_curriculum.replay_passes=0.25 +- ++datasets.vla_data.latency_curriculum.target_total_passes=2.0 +- ++datasets.vla_data.latency_curriculum.final_equalization=true +- ++datasets.vla_data.latency_curriculum.step_budget_mode=auto +- ++datasets.vla_data.latency_curriculum.eval_at_phase_end=false +- ++datasets.vla_data.latency_curriculum.save_at_phase_end=false +- ++datasets.vla_data.latency_curriculum.computed_plan=null +- ++datasets.vla_data.latency_curriculum.effective_batch_size=null +- ++datasets.vla_data.per_device_batch_size=4 +- ++datasets.vla_data.load_all_data_for_training=true +- ++datasets.vla_data.num_obs_frames=1 +- ++datasets.vla_data.image_mode=single +- ++datasets.vla_data.stitch_grid=[2,2] +- ++datasets.vla_data.obs_image_size=[224,224] +- ++datasets.vla_data.video_backend=torchvision_av +- ++datasets.vla_data.observation_indices=[-4,-3,-2,-1,0] +- ++datasets.vla_data.language_indices=[0] +- ++datasets.vla_data.state_indices=[0] +- ++datasets.vla_data.action_indices=[0,1,2,3,4,5,6,7] +- ++datasets.vla_data.pack_image_sequence=true +- ++datasets.vla_data.image_sequence_length=5 +- ++datasets.vla_data.context_images_column=observation.context_images +- ++dataset.source_hf= +- ++dataset.config_name=null +- ++dataset.source_subdir=null +- ++dataset.converted_name=flappy_train__bridge +- ++dataset.single_source_hf= +- ++dataset.mixed_source_hf= +- ++dataset.single_converted_name=flappy_train__bridge +- ++dataset.mixed_converted_name=flappy_mixed_latency_train +- ++dataset.single_latency_filter=null +- ++dataset.mixed_latency_filter=null +- ++dataset.force_download=false +- ++dataset.setup_force=false +- ++dataset.skip_verification=false +- ++dataset.verify_rows=200 +- ++dataset.max_episodes=null +- ++dataset.episodes_per_latency=null +- ++dataset.latency_filter=null +- ++dataset.debug_subset.enabled=false +- ++dataset.debug_subset.max_episodes=5 +- ++dataset.debug_subset.suffix=debug +- ++base_model.repo_id=Wan-AI/Wan2.2-TI2V-5B-Diffusers +- ++initialization.checkpoint_local_dir=playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1 +- ++initialization.checkpoint_hf_repo_id=StarVLA/WM4A-Wan2d2-OFT-LIBERO-4in1 +- ++initialization.checkpoint_filename=checkpoints/steps_60000_pytorch_model.pt +- ++trainer.max_train_steps=2000 +- ++trainer.num_warmup_steps=0 +- ++trainer.save_interval=0 +- ++trainer.eval_interval=500 +- ++trainer.eval_num_batches=50 +- ++trainer.per_latency_eval_num_batches=null +- ++trainer.eval_action_classification=false +- ++trainer.eval_action_classification_interval=null +- ++trainer.cc_f1_tolerance=1 +- ++trainer.learning_rate.base=2.5e-05 +- ++trainer.learning_rate.qwen_vl_interface=1e-05 +- ++trainer.learning_rate.action_model=0.0001 +- ++trainer.learning_rate.action_query_proj=0.0001 +- ++trainer.lr_scheduler_type=cosine_with_min_lr +- ++trainer.scheduler_specific_kwargs.min_lr=1e-06 +- ++trainer.freeze_modules= +- ++trainer.freeze_vit=false +- ++trainer.freeze_tied_embedding=false +- ++trainer.freeze_llm_layers=[] +- ++trainer.loss_scale.vla=1.0 +- ++trainer.loss_scale.vlm=0.1 +- ++trainer.max_grad_norm=1.0 +- ++trainer.weight_decay=0.0 +- ++trainer.logging_frequency=1 +- ++trainer.profile_timing.enabled=false +- ++trainer.profile_timing.log_interval=10 +- ++trainer.gradient_clipping=1.0 +- ++trainer.gradient_accumulation_steps=32 +- ++trainer.distributed_backend=none +- ++trainer.is_resume=false +- ++trainer.pretrained_checkpoint=null +- ++trainer.resume_step=0 +- ++trainer.reload_modules=null +- ++trainer.optimizer.name=AdamW +- ++trainer.optimizer.betas=[0.9,0.95] +- ++trainer.optimizer.eps=1e-08 +- ++trainer.optimizer.weight_decay=1e-08 +- ++trainer.optimizer.fused=true +- ++trainer.save_format=pt +- ++workspace_dir=WORKSPACE_DIR +- ++run_root_dir=results/Checkpoints +- ++seed=42 +- ++auth.env_file=null +- ++auth.hf_token_env=HF_TOKEN +- ++auth.wandb_api_key_env=WANDB_API_KEY +- ++paths.run_root_dir=results/Checkpoints +- ++paths.dataset_local_dir=data/flappy_fix_latency_1_200ep_context5 +- ++paths.dataset_cache_dir=null +- ++paths.base_model_dir=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers +- ++paths.accelerate_config=starVLA/config/deepseeds/deepspeed_zero2.yaml +- ++rl_games.model_alias=wan_oft +- ++rl_games.env_eval.image_size=224 +- ++rl_games.env_eval.frameskip=1 +- ++rl_games.env_eval.seed=42 +- ++rl_games.env_eval.fixed_episode_seeds=true +- ++rl_games.env_eval.latency_seed_stride=0 +- ++rl_games.env_eval.task_seed_stride=0 +- ++rl_games.env_eval.task_description= +- ++rl_games.env_eval.action_chunk_execution.enabled=false +- ++rl_games.env_eval.action_chunk_execution.chunk_size=null +- ++rl_games.env_eval.enabled=true +- ++rl_games.env_eval.distributed_mode=none +- ++rl_games.env_eval.vectorized.enabled=false +- ++rl_games.env_eval.vectorized.batch_size=1 +- ++rl_games.env_eval.latency.prompt_map_path=data/flappy_fix_latency_1_200ep_context5/flappy_train__bridge/latency_prompt_map_eval_0_4.json +- ++rl_games.env_eval.latency.mode=single +- ++rl_games.env_eval.latency.values=[0] +- ++rl_games.env_eval.mid_train.enabled=true +- ++rl_games.env_eval.mid_train.interval_steps=500 +- ++rl_games.env_eval.mid_train.latencies=[1] +- ++rl_games.env_eval.mid_train.num_episodes=5 +- ++rl_games.env_eval.mid_train.max_steps_per_episode=3600 +- ++rl_games.env_eval.post_train.enabled=true +- ++rl_games.env_eval.post_train.latencies=[0,1,2,3,4] +- ++rl_games.env_eval.post_train.num_episodes=20 +- ++rl_games.env_eval.post_train.max_steps_per_episode=3600 +- ++rl_games.task=flappy +- ++rl_games.initialization_mode=bridge +- ++rl_games.action_carrier=bridge +- ++checkpoint.load=none +- ++checkpoint.hf_repo_id=null +- ++checkpoint.save_best_model=false +- ++checkpoint.save_final_model=true +- ++checkpoint.save_pt_file=true +- ++checkpoint.save_training_state=false +- ++checkpoint.local.keep_last_n=1 +- ++checkpoint.sync.enabled=false +- ++checkpoint.sync.repo_id=null +- ++checkpoint.sync.keep_last_n=0 +- ++checkpoint.sync.sync_every_n_checkpoints=1 +- ++checkpoint.sync.resume_policy=local_latest +- ++run_id=wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce +- ++output_dir=null +- ++config_yaml=null +- ++is_debug=false +- ++version_id=0.21 +- ++run_root_dir=/workspace/starVLA/results/Checkpoints +- ++trainer.is_resume=false +- ++trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1/checkpoints/steps_60000_pytorch_model.pt +- ++trainer.resume_step=0 +- ++datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1_200ep_context5 +- ++framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers +- ++framework.world_model.base_wm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/train_starvla_hydra.log b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/train_starvla_hydra.log new file mode 100644 index 0000000000000000000000000000000000000000..2c8f1e15ca81831391dd9e508f2a8e5f1748d2b9 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/hydra/train_starvla_hydra.log @@ -0,0 +1,2078 @@ +[2026-06-29 14:49:33,764][starVLA.model.framework.share_tools][INFO] - [*] [apply_config_compat] normalised config from version_id=0.21 to '0.21' +[2026-06-29 14:49:33,764][starVLA.training.train_starvla][INFO] - VLA Training :: Warming Up +[2026-06-29 14:49:33,771][starVLA.training.train_starvla][INFO] - ✅ Configuration wrapped for access tracking +[2026-06-29 14:49:34,344][starVLA.model.modules.world_model.Wan2][INFO] - [*] Loading Wan2.2-TI2V from /workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers +[2026-06-29 14:49:45,928][starVLA.training.train_starvla][INFO] - Preloaded model checkpoint before Accelerator/DeepSpeed initialization: /workspace/starVLA/playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1/checkpoints/steps_60000_pytorch_model.pt +[2026-06-29 14:49:45,944][starVLA.training.train_starvla][INFO] - Creating VLA Dataset with Mixture `flappy_train__bridge` +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from Behavior (data_config): ['BEHAVIOR_challenge'] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from DOMINO: ['robotwin'] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from DOMINO (data_config): [] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from DOMINO (data_config): ['domino', 'domino_clean', 'domino_random', 'domino_cotrain'] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from Franka: ['custom_robot_config', 'demo_sim_franka_delta_joints', 'SO101'] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from Franka (data_config): [] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from Franka (data_config): ['custom_dataset', 'custom_dataset_2', 'demo_sim_pick_place', 'SO101_pick'] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from LIBERO: ['libero_franka'] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from LIBERO (data_config): [] +[2026-06-29 14:49:46,238][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from LIBERO (data_config): ['libero_all', 'libero_goal', 'multi_robot'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from RoboChallenge_table30v2: ['ur5_robochallenge', 'arx5_robochallenge', 'dosw1_robochallenge'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from RoboChallenge_table30v2 (data_config): ['ur5_robochallenge', 'arx5_robochallenge', 'dosw1_robochallenge'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from RoboChallenge_table30v2 (data_config): ['robochallenge_table30v2_shred_paper', 'robochallenge_table30v2_ur5_all', 'robochallenge_table30v2_arx5_all', 'robochallenge_table30v2_dosw1_all'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from Robocasa_365: ['panda_omron_robocasa365'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from Robocasa_365 (data_config): [] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from Robocasa_365 (data_config): ['robocasa365_open_drawer_target_human', 'robocasa365_atomic_target_human_all', 'robocasa365_composite_target_human_all', 'robocasa365_target_human_all'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from Robocasa_tabletop: ['fourier_gr1_arms_waist'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from Robocasa_tabletop (data_config): [] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from Robocasa_tabletop (data_config): ['fourier_gr1_unified_1000'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from Robotwin: ['robotwin', 'robotwin50', 'arx_x5'] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from Robotwin (data_config): [] +[2026-06-29 14:49:46,239][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from Robotwin (data_config): ['robotwin_all', 'robotwin_all_50', 'robotwin', 'robotwin_task1', 'robotwin_task2', 'arx_x5'] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from SimplerEnv: ['oxe_droid', 'oxe_bridge', 'oxe_rt1'] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from SimplerEnv (data_config): [] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from SimplerEnv (data_config): ['bridge', 'bridge_rt_1'] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from VLA-Arena: ['vla_arena_franka'] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from VLA-Arena (data_config): [] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from VLA-Arena (data_config): ['vla_arena_L0_S', 'vla_arena_L0_M', 'vla_arena_L0_L'] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded data_config from rl_games: ['rl_games_flappy', 'rl_games_demon_attack', 'rl_games_deadly_corridor'] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded embodiment_tags from rl_games (data_config): ['rl_games_flappy', 'rl_games_demon_attack', 'rl_games_deadly_corridor'] +[2026-06-29 14:49:46,240][starVLA.dataloader.gr00t_lerobot.registry][INFO] - [registry] Loaded mixtures from rl_games (data_config): ['flappy_train', 'flappy_train__bridge', 'flappy_mixed_latency_train', 'flappy_mixed_latency_train__bridge', 'demon_attack_train', 'demon_attack_train__bridge', 'demon_attack_mixed_latency_train', 'demon_attack_mixed_latency_train__bridge', 'deadly_corridor_train', 'deadly_corridor_train__bridge', 'deadly_corridor_mixed_latency_train', 'deadly_corridor_mixed_latency_train__bridge'] +[2026-06-29 14:50:06,436][starVLA.training.train_starvla][INFO] - Creating VLA Eval Dataset with Mixture `flappy_train__bridge__val` +[2026-06-29 14:50:10,727][starVLA.training.train_starvla][INFO] - 📝 Full config saved at /workspace/starVLA/results/Checkpoints/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.full.yaml +[2026-06-29 14:50:10,731][starVLA.training.train_starvla][INFO] - 📊 Accessed config snapshot saved at /workspace/starVLA/results/Checkpoints/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/config.yaml +[2026-06-29 14:50:10,731][starVLA.training.train_starvla][INFO] - No pretrained checkpoint provided. Starting training from scratch. +[2026-06-29 14:50:16,596][starVLA.training.train_starvla][INFO] - ***** Training Configuration ***** +[2026-06-29 14:50:16,596][starVLA.training.train_starvla][INFO] - Total optimization steps = 2000 +[2026-06-29 14:50:16,596][starVLA.training.train_starvla][INFO] - Per device batch size = 4 +[2026-06-29 14:50:16,596][starVLA.training.train_starvla][INFO] - Gradient accumulation steps = 32 +[2026-06-29 14:50:16,596][starVLA.training.train_starvla][INFO] - Total batch size = 128 +[2026-06-29 14:51:09,960][starVLA.training.train_starvla][INFO] - Step 1, Loss: {'train/loss': 0.7087021470069885, 'train/grad_norm_pre_clip': 18.822635650634766, 'timing/data': 0.0003733038902282715, 'timing/model': 1.2268622650299221, 'learning_rate/action_model': 9.999994078238577e-05, 'learning_rate/action_query_proj': 9.999994078238577e-05, 'learning_rate/base': 2.4999985195596442e-05, 'epoch': 0.0}) +[2026-06-29 14:51:42,092][starVLA.training.train_starvla][INFO] - Step 2, Loss: {'train/loss': 0.2893279492855072, 'train/grad_norm_pre_clip': 0.668279767036438, 'timing/data': 0.00022906181402504444, 'timing/model': 1.2133935478050262, 'learning_rate/action_model': 9.99997631296892e-05, 'learning_rate/action_query_proj': 9.99997631296892e-05, 'learning_rate/base': 2.49999407824223e-05, 'epoch': 0.0}) +[2026-06-29 14:52:14,337][starVLA.training.train_starvla][INFO] - Step 3, Loss: {'train/loss': 0.5872522592544556, 'train/grad_norm_pre_clip': 11.599517822265625, 'timing/data': 0.00019632093608379364, 'timing/model': 1.2170117769856006, 'learning_rate/action_model': 9.999946704234862e-05, 'learning_rate/action_query_proj': 9.999946704234862e-05, 'learning_rate/base': 2.4999866760587155e-05, 'epoch': 0.0}) +[2026-06-29 14:52:46,590][starVLA.training.train_starvla][INFO] - Step 4, Loss: {'train/loss': 0.2902737557888031, 'train/grad_norm_pre_clip': 1.6435718536376953, 'timing/data': 0.00018959795124828815, 'timing/model': 1.2201841019559652, 'learning_rate/action_model': 9.999905252109459e-05, 'learning_rate/action_query_proj': 9.999905252109459e-05, 'learning_rate/base': 2.4999763130273646e-05, 'epoch': 0.0}) +[2026-06-29 14:53:18,899][starVLA.training.train_starvla][INFO] - Step 5, Loss: {'train/loss': 0.7365295886993408, 'train/grad_norm_pre_clip': 6.079817771911621, 'timing/data': 0.0002447508741170168, 'timing/model': 1.2165986089967191, 'learning_rate/action_model': 9.999851956694992e-05, 'learning_rate/action_query_proj': 9.999851956694992e-05, 'learning_rate/base': 2.499962989173748e-05, 'epoch': 0.0}) +[2026-06-29 14:53:51,254][starVLA.training.train_starvla][INFO] - Step 6, Loss: {'train/loss': 0.9493237137794495, 'train/grad_norm_pre_clip': 3.356424331665039, 'timing/data': 0.00019925599917769432, 'timing/model': 1.2238455251790583, 'learning_rate/action_model': 9.99978681812296e-05, 'learning_rate/action_query_proj': 9.99978681812296e-05, 'learning_rate/base': 2.49994670453074e-05, 'epoch': 0.0}) +[2026-06-29 14:54:23,621][starVLA.training.train_starvla][INFO] - Step 7, Loss: {'train/loss': 0.7313209772109985, 'train/grad_norm_pre_clip': 5.33993673324585, 'timing/data': 0.00020053819753229618, 'timing/model': 1.2232642159797251, 'learning_rate/action_model': 9.999709836554088e-05, 'learning_rate/action_query_proj': 9.999709836554088e-05, 'learning_rate/base': 2.499927459138522e-05, 'epoch': 0.0}) +[2026-06-29 14:54:55,945][starVLA.training.train_starvla][INFO] - Step 8, Loss: {'train/loss': 0.8606191873550415, 'train/grad_norm_pre_clip': 2.2490572929382324, 'timing/data': 0.00019322405569255352, 'timing/model': 1.2233491679653525, 'learning_rate/action_model': 9.999621012178318e-05, 'learning_rate/action_query_proj': 9.999621012178318e-05, 'learning_rate/base': 2.4999052530445796e-05, 'epoch': 0.0}) +[2026-06-29 14:55:28,305][starVLA.training.train_starvla][INFO] - Step 9, Loss: {'train/loss': 0.873947262763977, 'train/grad_norm_pre_clip': 1.3855791091918945, 'timing/data': 0.00020680995658040047, 'timing/model': 1.221255716867745, 'learning_rate/action_model': 9.999520345214816e-05, 'learning_rate/action_query_proj': 9.999520345214816e-05, 'learning_rate/base': 2.499880086303704e-05, 'epoch': 0.0}) +[2026-06-29 14:56:00,677][starVLA.training.train_starvla][INFO] - Step 10, Loss: {'train/loss': 0.3375788927078247, 'train/grad_norm_pre_clip': 4.770168781280518, 'timing/data': 0.0003071089740842581, 'timing/model': 1.218357764184475, 'learning_rate/action_model': 9.999407835911971e-05, 'learning_rate/action_query_proj': 9.999407835911971e-05, 'learning_rate/base': 2.4998519589779927e-05, 'epoch': 0.0}) +[2026-06-29 14:56:33,021][starVLA.training.train_starvla][INFO] - Step 11, Loss: {'train/loss': 0.5481567978858948, 'train/grad_norm_pre_clip': 1.137817621231079, 'timing/data': 0.00023614591918885708, 'timing/model': 1.2183544200379401, 'learning_rate/action_model': 9.999283484547385e-05, 'learning_rate/action_query_proj': 9.999283484547385e-05, 'learning_rate/base': 2.4998208711368463e-05, 'epoch': 0.0}) +[2026-06-29 14:57:05,399][starVLA.training.train_starvla][INFO] - Step 12, Loss: {'train/loss': 0.37409186363220215, 'train/grad_norm_pre_clip': 2.7442421913146973, 'timing/data': 0.0002593791577965021, 'timing/model': 1.210368013009429, 'learning_rate/action_model': 9.999147291427884e-05, 'learning_rate/action_query_proj': 9.999147291427884e-05, 'learning_rate/base': 2.499786822856971e-05, 'epoch': 0.01}) +[2026-06-29 14:57:37,694][starVLA.training.train_starvla][INFO] - Step 13, Loss: {'train/loss': 0.4985688328742981, 'train/grad_norm_pre_clip': 3.313009023666382, 'timing/data': 0.00022890092805027962, 'timing/model': 1.2192852329462767, 'learning_rate/action_model': 9.99899925688951e-05, 'learning_rate/action_query_proj': 9.99899925688951e-05, 'learning_rate/base': 2.4997498142223774e-05, 'epoch': 0.01}) +[2026-06-29 14:58:10,052][starVLA.training.train_starvla][INFO] - Step 14, Loss: {'train/loss': 0.6107640862464905, 'train/grad_norm_pre_clip': 5.978645324707031, 'timing/data': 0.00025020097382366657, 'timing/model': 1.2185359841678292, 'learning_rate/action_model': 9.998839381297525e-05, 'learning_rate/action_query_proj': 9.998839381297525e-05, 'learning_rate/base': 2.4997098453243813e-05, 'epoch': 0.01}) +[2026-06-29 14:58:42,388][starVLA.training.train_starvla][INFO] - Step 15, Loss: {'train/loss': 0.0994676724076271, 'train/grad_norm_pre_clip': 2.04585862159729, 'timing/data': 0.00019942596554756165, 'timing/model': 1.2242735649924725, 'learning_rate/action_model': 9.998667665046403e-05, 'learning_rate/action_query_proj': 9.998667665046403e-05, 'learning_rate/base': 2.4996669162616007e-05, 'epoch': 0.01}) +[2026-06-29 14:59:14,705][starVLA.training.train_starvla][INFO] - Step 16, Loss: {'train/loss': 0.4849071204662323, 'train/grad_norm_pre_clip': 7.146388053894043, 'timing/data': 0.0002956369426101446, 'timing/model': 1.2209325160365552, 'learning_rate/action_model': 9.99848410855984e-05, 'learning_rate/action_query_proj': 9.99848410855984e-05, 'learning_rate/base': 2.49962102713996e-05, 'epoch': 0.01}) +[2026-06-29 14:59:47,079][starVLA.training.train_starvla][INFO] - Step 17, Loss: {'train/loss': 0.3699740767478943, 'train/grad_norm_pre_clip': 4.3915791511535645, 'timing/data': 0.00021391408517956734, 'timing/model': 1.2238150101620704, 'learning_rate/action_model': 9.998288712290742e-05, 'learning_rate/action_query_proj': 9.998288712290742e-05, 'learning_rate/base': 2.4995721780726855e-05, 'epoch': 0.01}) +[2026-06-29 15:00:19,533][starVLA.training.train_starvla][INFO] - Step 18, Loss: {'train/loss': 0.31613028049468994, 'train/grad_norm_pre_clip': 1.8727785348892212, 'timing/data': 0.0003054160624742508, 'timing/model': 1.218435934977606, 'learning_rate/action_model': 9.998081476721232e-05, 'learning_rate/action_query_proj': 9.998081476721232e-05, 'learning_rate/base': 2.499520369180308e-05, 'epoch': 0.01}) +[2026-06-29 15:00:51,865][starVLA.training.train_starvla][INFO] - Step 19, Loss: {'train/loss': 0.33953896164894104, 'train/grad_norm_pre_clip': 2.3796160221099854, 'timing/data': 0.0002935829106718302, 'timing/model': 1.2239559160079807, 'learning_rate/action_model': 9.99786240236264e-05, 'learning_rate/action_query_proj': 9.99786240236264e-05, 'learning_rate/base': 2.49946560059066e-05, 'epoch': 0.01}) +[2026-06-29 15:01:24,223][starVLA.training.train_starvla][INFO] - Step 20, Loss: {'train/loss': 0.2993757426738739, 'train/grad_norm_pre_clip': 2.375732183456421, 'timing/data': 0.00020230095833539963, 'timing/model': 1.224797380156815, 'learning_rate/action_model': 9.997631489755513e-05, 'learning_rate/action_query_proj': 9.997631489755513e-05, 'learning_rate/base': 2.4994078724388782e-05, 'epoch': 0.01}) +[2026-06-29 15:01:56,532][starVLA.training.train_starvla][INFO] - Step 21, Loss: {'train/loss': 0.14540185034275055, 'train/grad_norm_pre_clip': 3.8413069248199463, 'timing/data': 0.00033838790841400623, 'timing/model': 1.21879344410263, 'learning_rate/action_model': 9.997388739469601e-05, 'learning_rate/action_query_proj': 9.997388739469601e-05, 'learning_rate/base': 2.4993471848674004e-05, 'epoch': 0.01}) +[2026-06-29 15:02:28,891][starVLA.training.train_starvla][INFO] - Step 22, Loss: {'train/loss': 0.15927034616470337, 'train/grad_norm_pre_clip': 1.780666708946228, 'timing/data': 0.0002490789629518986, 'timing/model': 1.2174050381872803, 'learning_rate/action_model': 9.997134152103871e-05, 'learning_rate/action_query_proj': 9.997134152103871e-05, 'learning_rate/base': 2.4992835380259678e-05, 'epoch': 0.01}) +[2026-06-29 15:03:01,235][starVLA.training.train_starvla][INFO] - Step 23, Loss: {'train/loss': 0.4085595905780792, 'train/grad_norm_pre_clip': 1.1609247922897339, 'timing/data': 0.00019637984223663807, 'timing/model': 1.2209535189904273, 'learning_rate/action_model': 9.996867728286489e-05, 'learning_rate/action_query_proj': 9.996867728286489e-05, 'learning_rate/base': 2.499216932071622e-05, 'epoch': 0.01}) +[2026-06-29 15:03:33,593][starVLA.training.train_starvla][INFO] - Step 24, Loss: {'train/loss': 0.188778817653656, 'train/grad_norm_pre_clip': 8.067625045776367, 'timing/data': 0.00027160183526575565, 'timing/model': 1.2227902768645436, 'learning_rate/action_model': 9.996589468674828e-05, 'learning_rate/action_query_proj': 9.996589468674828e-05, 'learning_rate/base': 2.499147367168707e-05, 'epoch': 0.01}) +[2026-06-29 15:04:05,978][starVLA.training.train_starvla][INFO] - Step 25, Loss: {'train/loss': 0.12968093156814575, 'train/grad_norm_pre_clip': 6.495263576507568, 'timing/data': 0.0002233409322798252, 'timing/model': 1.223862234968692, 'learning_rate/action_model': 9.996299373955472e-05, 'learning_rate/action_query_proj': 9.996299373955472e-05, 'learning_rate/base': 2.499074843488868e-05, 'epoch': 0.01}) +[2026-06-29 15:04:38,361][starVLA.training.train_starvla][INFO] - Step 26, Loss: {'train/loss': 0.5186768174171448, 'train/grad_norm_pre_clip': 2.4854791164398193, 'timing/data': 0.0002686958760023117, 'timing/model': 1.220274078892544, 'learning_rate/action_model': 9.995997444844192e-05, 'learning_rate/action_query_proj': 9.995997444844192e-05, 'learning_rate/base': 2.498999361211048e-05, 'epoch': 0.01}) +[2026-06-29 15:05:10,703][starVLA.training.train_starvla][INFO] - Step 27, Loss: {'train/loss': 0.2203730344772339, 'train/grad_norm_pre_clip': 8.141965866088867, 'timing/data': 0.0004007960669696331, 'timing/model': 1.2213744828477502, 'learning_rate/action_model': 9.995683682085976e-05, 'learning_rate/action_query_proj': 9.995683682085976e-05, 'learning_rate/base': 2.498920920521494e-05, 'epoch': 0.01}) +[2026-06-29 15:05:43,046][starVLA.training.train_starvla][INFO] - Step 28, Loss: {'train/loss': 0.1847989857196808, 'train/grad_norm_pre_clip': 6.947931289672852, 'timing/data': 0.000206309137865901, 'timing/model': 1.2245121460873634, 'learning_rate/action_model': 9.995358086455e-05, 'learning_rate/action_query_proj': 9.995358086455e-05, 'learning_rate/base': 2.49883952161375e-05, 'epoch': 0.01}) +[2026-06-29 15:06:15,388][starVLA.training.train_starvla][INFO] - Step 29, Loss: {'train/loss': 0.8100578188896179, 'train/grad_norm_pre_clip': 6.920132637023926, 'timing/data': 0.0002034630160778761, 'timing/model': 1.221998750930652, 'learning_rate/action_model': 9.995020658754636e-05, 'learning_rate/action_query_proj': 9.995020658754636e-05, 'learning_rate/base': 2.498755164688659e-05, 'epoch': 0.01}) +[2026-06-29 15:06:47,725][starVLA.training.train_starvla][INFO] - Step 30, Loss: {'train/loss': 0.379520446062088, 'train/grad_norm_pre_clip': 2.1431467533111572, 'timing/data': 0.00021151895634829998, 'timing/model': 1.2183170360513031, 'learning_rate/action_model': 9.994671399817457e-05, 'learning_rate/action_query_proj': 9.994671399817457e-05, 'learning_rate/base': 2.4986678499543643e-05, 'epoch': 0.01}) +[2026-06-29 15:07:20,312][starVLA.training.train_starvla][INFO] - Step 31, Loss: {'train/loss': 0.12420755624771118, 'train/grad_norm_pre_clip': 3.889509916305542, 'timing/data': 0.00019152089953422546, 'timing/model': 1.2195567768067122, 'learning_rate/action_model': 9.994310310505222e-05, 'learning_rate/action_query_proj': 9.994310310505222e-05, 'learning_rate/base': 2.4985775776263055e-05, 'epoch': 0.01}) +[2026-06-29 15:07:52,633][starVLA.training.train_starvla][INFO] - Step 32, Loss: {'train/loss': 0.22082781791687012, 'train/grad_norm_pre_clip': 4.980716228485107, 'timing/data': 0.00018895603716373444, 'timing/model': 1.2163284979760647, 'learning_rate/action_model': 9.993937391708885e-05, 'learning_rate/action_query_proj': 9.993937391708885e-05, 'learning_rate/base': 2.4984843479272213e-05, 'epoch': 0.01}) +[2026-06-29 15:08:24,994][starVLA.training.train_starvla][INFO] - Step 33, Loss: {'train/loss': 0.00748074147850275, 'train/grad_norm_pre_clip': 12.16554069519043, 'timing/data': 0.00020365300588309765, 'timing/model': 1.2220083680003881, 'learning_rate/action_model': 9.993552644348583e-05, 'learning_rate/action_query_proj': 9.993552644348583e-05, 'learning_rate/base': 2.498388161087146e-05, 'epoch': 0.01}) +[2026-06-29 15:08:57,315][starVLA.training.train_starvla][INFO] - Step 34, Loss: {'train/loss': 0.2681119441986084, 'train/grad_norm_pre_clip': 8.281848907470703, 'timing/data': 0.00025862688198685646, 'timing/model': 1.2193060209974647, 'learning_rate/action_model': 9.993156069373646e-05, 'learning_rate/action_query_proj': 9.993156069373646e-05, 'learning_rate/base': 2.4982890173434116e-05, 'epoch': 0.01}) +[2026-06-29 15:09:29,652][starVLA.training.train_starvla][INFO] - Step 35, Loss: {'train/loss': 0.13438284397125244, 'train/grad_norm_pre_clip': 5.782277584075928, 'timing/data': 0.00033881794661283493, 'timing/model': 1.2226624339818954, 'learning_rate/action_model': 9.992747667762583e-05, 'learning_rate/action_query_proj': 9.992747667762583e-05, 'learning_rate/base': 2.4981869169406458e-05, 'epoch': 0.02}) +[2026-06-29 15:10:01,994][starVLA.training.train_starvla][INFO] - Step 36, Loss: {'train/loss': 0.30955779552459717, 'train/grad_norm_pre_clip': 5.230932712554932, 'timing/data': 0.00019511696882545948, 'timing/model': 1.2250757610891014, 'learning_rate/action_model': 9.992327440523081e-05, 'learning_rate/action_query_proj': 9.992327440523081e-05, 'learning_rate/base': 2.4980818601307702e-05, 'epoch': 0.02}) +[2026-06-29 15:10:34,339][starVLA.training.train_starvla][INFO] - Step 37, Loss: {'train/loss': 0.3104856610298157, 'train/grad_norm_pre_clip': 3.013286828994751, 'timing/data': 0.00019323406741023064, 'timing/model': 1.2199388099834323, 'learning_rate/action_model': 9.99189538869201e-05, 'learning_rate/action_query_proj': 9.99189538869201e-05, 'learning_rate/base': 2.4979738471730024e-05, 'epoch': 0.02}) +[2026-06-29 15:11:06,676][starVLA.training.train_starvla][INFO] - Step 38, Loss: {'train/loss': 0.30442702770233154, 'train/grad_norm_pre_clip': 9.320265769958496, 'timing/data': 0.0002830531448125839, 'timing/model': 1.2180656290147454, 'learning_rate/action_model': 9.991451513335418e-05, 'learning_rate/action_query_proj': 9.991451513335418e-05, 'learning_rate/base': 2.4978628783338545e-05, 'epoch': 0.02}) +[2026-06-29 15:11:38,999][starVLA.training.train_starvla][INFO] - Step 39, Loss: {'train/loss': 0.292712539434433, 'train/grad_norm_pre_clip': 2.466881275177002, 'timing/data': 0.0002451320178806782, 'timing/model': 1.2252388070337474, 'learning_rate/action_model': 9.990995815548518e-05, 'learning_rate/action_query_proj': 9.990995815548518e-05, 'learning_rate/base': 2.4977489538871295e-05, 'epoch': 0.02}) +[2026-06-29 15:12:11,293][starVLA.training.train_starvla][INFO] - Step 40, Loss: {'train/loss': 0.3030918836593628, 'train/grad_norm_pre_clip': 2.2065324783325195, 'timing/data': 0.00019052904099225998, 'timing/model': 1.2180274480488151, 'learning_rate/action_model': 9.990528296455703e-05, 'learning_rate/action_query_proj': 9.990528296455703e-05, 'learning_rate/base': 2.4976320741139258e-05, 'epoch': 0.02}) +[2026-06-29 15:12:43,635][starVLA.training.train_starvla][INFO] - Step 41, Loss: {'train/loss': 0.409098744392395, 'train/grad_norm_pre_clip': 4.5960307121276855, 'timing/data': 0.0002496500965207815, 'timing/model': 1.2158563290722668, 'learning_rate/action_model': 9.99004895721053e-05, 'learning_rate/action_query_proj': 9.99004895721053e-05, 'learning_rate/base': 2.4975122393026324e-05, 'epoch': 0.02}) +[2026-06-29 15:13:15,938][starVLA.training.train_starvla][INFO] - Step 42, Loss: {'train/loss': 0.11483480036258698, 'train/grad_norm_pre_clip': 2.0440406799316406, 'timing/data': 0.00019009900279343128, 'timing/model': 1.2199942080769688, 'learning_rate/action_model': 9.98955779899572e-05, 'learning_rate/action_query_proj': 9.98955779899572e-05, 'learning_rate/base': 2.49738944974893e-05, 'epoch': 0.02}) +[2026-06-29 15:13:48,215][starVLA.training.train_starvla][INFO] - Step 43, Loss: {'train/loss': 0.80901700258255, 'train/grad_norm_pre_clip': 5.763251781463623, 'timing/data': 0.00022505386732518673, 'timing/model': 1.2193718790076673, 'learning_rate/action_model': 9.989054823023156e-05, 'learning_rate/action_query_proj': 9.989054823023156e-05, 'learning_rate/base': 2.497263705755789e-05, 'epoch': 0.02}) +[2026-06-29 15:14:20,562][starVLA.training.train_starvla][INFO] - Step 44, Loss: {'train/loss': 0.04744620993733406, 'train/grad_norm_pre_clip': 9.015772819519043, 'timing/data': 0.00019362405873835087, 'timing/model': 1.218709924025461, 'learning_rate/action_model': 9.98854003053388e-05, 'learning_rate/action_query_proj': 9.98854003053388e-05, 'learning_rate/base': 2.49713500763347e-05, 'epoch': 0.02}) +[2026-06-29 15:14:52,893][starVLA.training.train_starvla][INFO] - Step 45, Loss: {'train/loss': 0.011488119140267372, 'train/grad_norm_pre_clip': 2.896881341934204, 'timing/data': 0.0002908890601247549, 'timing/model': 1.2178118049632758, 'learning_rate/action_model': 9.988013422798097e-05, 'learning_rate/action_query_proj': 9.988013422798097e-05, 'learning_rate/base': 2.4970033556995243e-05, 'epoch': 0.02}) +[2026-06-29 15:15:25,202][starVLA.training.train_starvla][INFO] - Step 46, Loss: {'train/loss': 0.03455426171422005, 'train/grad_norm_pre_clip': 1.6130256652832031, 'timing/data': 0.00025529111735522747, 'timing/model': 1.2222875540610403, 'learning_rate/action_model': 9.987475001115155e-05, 'learning_rate/action_query_proj': 9.987475001115155e-05, 'learning_rate/base': 2.4968687502787886e-05, 'epoch': 0.02}) +[2026-06-29 15:15:57,560][starVLA.training.train_starvla][INFO] - Step 47, Loss: {'train/loss': 0.03574540466070175, 'train/grad_norm_pre_clip': 7.8773980140686035, 'timing/data': 0.00018601003102958202, 'timing/model': 1.219730188138783, 'learning_rate/action_model': 9.986924766813554e-05, 'learning_rate/action_query_proj': 9.986924766813554e-05, 'learning_rate/base': 2.4967311917033885e-05, 'epoch': 0.02}) +[2026-06-29 15:16:29,928][starVLA.training.train_starvla][INFO] - Step 48, Loss: {'train/loss': 0.06761551648378372, 'train/grad_norm_pre_clip': 3.7165777683258057, 'timing/data': 0.00018600095063447952, 'timing/model': 1.220360861159861, 'learning_rate/action_model': 9.986362721250947e-05, 'learning_rate/action_query_proj': 9.986362721250947e-05, 'learning_rate/base': 2.4965906803127368e-05, 'epoch': 0.02}) +[2026-06-29 15:17:02,285][starVLA.training.train_starvla][INFO] - Step 49, Loss: {'train/loss': 0.660344123840332, 'train/grad_norm_pre_clip': 7.8617095947265625, 'timing/data': 0.0003287999425083399, 'timing/model': 1.223009407054633, 'learning_rate/action_model': 9.985788865814124e-05, 'learning_rate/action_query_proj': 9.985788865814124e-05, 'learning_rate/base': 2.496447216453531e-05, 'epoch': 0.02}) +[2026-06-29 15:17:34,833][starVLA.training.train_starvla][INFO] - Step 50, Loss: {'train/loss': 0.08157990127801895, 'train/grad_norm_pre_clip': 2.2137489318847656, 'timing/data': 0.00021748105064034462, 'timing/model': 1.21933064609766, 'learning_rate/action_model': 9.985203201919016e-05, 'learning_rate/action_query_proj': 9.985203201919016e-05, 'learning_rate/base': 2.496300800479754e-05, 'epoch': 0.02}) +[2026-06-29 15:18:07,161][starVLA.training.train_starvla][INFO] - Step 51, Loss: {'train/loss': 0.06519419699907303, 'train/grad_norm_pre_clip': 2.6188926696777344, 'timing/data': 0.00029218290001153946, 'timing/model': 1.2168747619725764, 'learning_rate/action_model': 9.984605731010687e-05, 'learning_rate/action_query_proj': 9.984605731010687e-05, 'learning_rate/base': 2.496151432752672e-05, 'epoch': 0.02}) +[2026-06-29 15:18:39,508][starVLA.training.train_starvla][INFO] - Step 52, Loss: {'train/loss': 1.0680301189422607, 'train/grad_norm_pre_clip': 5.978086471557617, 'timing/data': 0.0003557121381163597, 'timing/model': 1.2205937278922647, 'learning_rate/action_model': 9.983996454563343e-05, 'learning_rate/action_query_proj': 9.983996454563343e-05, 'learning_rate/base': 2.4959991136408358e-05, 'epoch': 0.02}) +[2026-06-29 15:19:11,835][starVLA.training.train_starvla][INFO] - Step 53, Loss: {'train/loss': 0.15554440021514893, 'train/grad_norm_pre_clip': 11.4601469039917, 'timing/data': 0.00028255395591259, 'timing/model': 1.2227321472018957, 'learning_rate/action_model': 9.983375374080311e-05, 'learning_rate/action_query_proj': 9.983375374080311e-05, 'learning_rate/base': 2.4958438435200777e-05, 'epoch': 0.02}) +[2026-06-29 15:19:44,176][starVLA.training.train_starvla][INFO] - Step 54, Loss: {'train/loss': 0.15088501572608948, 'train/grad_norm_pre_clip': 13.286211967468262, 'timing/data': 0.00026456895284354687, 'timing/model': 1.2247277330607176, 'learning_rate/action_model': 9.982742491094044e-05, 'learning_rate/action_query_proj': 9.982742491094044e-05, 'learning_rate/base': 2.495685622773511e-05, 'epoch': 0.02}) +[2026-06-29 15:20:16,505][starVLA.training.train_starvla][INFO] - Step 55, Loss: {'train/loss': 0.6026718616485596, 'train/grad_norm_pre_clip': 7.978915214538574, 'timing/data': 0.0001907101832330227, 'timing/model': 1.217455544974655, 'learning_rate/action_model': 9.982097807166118e-05, 'learning_rate/action_query_proj': 9.982097807166118e-05, 'learning_rate/base': 2.4955244517915295e-05, 'epoch': 0.02}) +[2026-06-29 15:20:48,852][starVLA.training.train_starvla][INFO] - Step 56, Loss: {'train/loss': 0.040169306099414825, 'train/grad_norm_pre_clip': 4.738067150115967, 'timing/data': 0.0004040030762553215, 'timing/model': 1.218803864903748, 'learning_rate/action_model': 9.981441323887228e-05, 'learning_rate/action_query_proj': 9.981441323887228e-05, 'learning_rate/base': 2.495360330971807e-05, 'epoch': 0.02}) +[2026-06-29 15:21:21,273][starVLA.training.train_starvla][INFO] - Step 57, Loss: {'train/loss': 0.24962419271469116, 'train/grad_norm_pre_clip': 2.447364330291748, 'timing/data': 0.00019020913168787956, 'timing/model': 1.220842001028359, 'learning_rate/action_model': 9.98077304287718e-05, 'learning_rate/action_query_proj': 9.98077304287718e-05, 'learning_rate/base': 2.495193260719295e-05, 'epoch': 0.02}) +[2026-06-29 15:21:53,623][starVLA.training.train_starvla][INFO] - Step 58, Loss: {'train/loss': 0.4338676631450653, 'train/grad_norm_pre_clip': 6.881341457366943, 'timing/data': 0.00019070995040237904, 'timing/model': 1.2214140240103006, 'learning_rate/action_model': 9.980092965784891e-05, 'learning_rate/action_query_proj': 9.980092965784891e-05, 'learning_rate/base': 2.4950232414462228e-05, 'epoch': 0.02}) +[2026-06-29 15:22:25,927][starVLA.training.train_starvla][INFO] - Step 59, Loss: {'train/loss': 0.30748337507247925, 'train/grad_norm_pre_clip': 7.383453369140625, 'timing/data': 0.0002900068648159504, 'timing/model': 1.2202688611578196, 'learning_rate/action_model': 9.979401094288385e-05, 'learning_rate/action_query_proj': 9.979401094288385e-05, 'learning_rate/base': 2.4948502735720962e-05, 'epoch': 0.03}) +[2026-06-29 15:22:58,265][starVLA.training.train_starvla][INFO] - Step 60, Loss: {'train/loss': 0.5146402716636658, 'train/grad_norm_pre_clip': 8.214204788208008, 'timing/data': 0.0001856300514191389, 'timing/model': 1.2225582210812718, 'learning_rate/action_model': 9.978697430094785e-05, 'learning_rate/action_query_proj': 9.978697430094785e-05, 'learning_rate/base': 2.494674357523696e-05, 'epoch': 0.03}) +[2026-06-29 15:23:30,598][starVLA.training.train_starvla][INFO] - Step 61, Loss: {'train/loss': 0.02867106720805168, 'train/grad_norm_pre_clip': 6.350340366363525, 'timing/data': 0.00028839404694736004, 'timing/model': 1.2196547340136021, 'learning_rate/action_model': 9.977981974940312e-05, 'learning_rate/action_query_proj': 9.977981974940312e-05, 'learning_rate/base': 2.494495493735078e-05, 'epoch': 0.03}) +[2026-06-29 15:24:02,883][starVLA.training.train_starvla][INFO] - Step 62, Loss: {'train/loss': 0.18115057051181793, 'train/grad_norm_pre_clip': 11.086596488952637, 'timing/data': 0.0003747569862753153, 'timing/model': 1.233257720945403, 'learning_rate/action_model': 9.977254730590281e-05, 'learning_rate/action_query_proj': 9.977254730590281e-05, 'learning_rate/base': 2.4943136826475702e-05, 'epoch': 0.03}) +[2026-06-29 15:24:35,214][starVLA.training.train_starvla][INFO] - Step 63, Loss: {'train/loss': 0.002737222472205758, 'train/grad_norm_pre_clip': 3.510255813598633, 'timing/data': 0.00019322498701512814, 'timing/model': 1.2204429900739342, 'learning_rate/action_model': 9.976515698839095e-05, 'learning_rate/action_query_proj': 9.976515698839095e-05, 'learning_rate/base': 2.4941289247097737e-05, 'epoch': 0.03}) +[2026-06-29 15:25:07,584][starVLA.training.train_starvla][INFO] - Step 64, Loss: {'train/loss': 0.08474253863096237, 'train/grad_norm_pre_clip': 3.5424342155456543, 'timing/data': 0.0001908009871840477, 'timing/model': 1.2185095469467342, 'learning_rate/action_model': 9.975764881510242e-05, 'learning_rate/action_query_proj': 9.975764881510242e-05, 'learning_rate/base': 2.4939412203775605e-05, 'epoch': 0.03}) +[2026-06-29 15:25:39,928][starVLA.training.train_starvla][INFO] - Step 65, Loss: {'train/loss': 0.2903166115283966, 'train/grad_norm_pre_clip': 3.665071964263916, 'timing/data': 0.0002545102033764124, 'timing/model': 1.2215184699743986, 'learning_rate/action_model': 9.975002280456287e-05, 'learning_rate/action_query_proj': 9.975002280456287e-05, 'learning_rate/base': 2.4937505701140718e-05, 'epoch': 0.03}) +[2026-06-29 15:26:12,259][starVLA.training.train_starvla][INFO] - Step 66, Loss: {'train/loss': 0.2577311098575592, 'train/grad_norm_pre_clip': 4.922474384307861, 'timing/data': 0.0003617431502789259, 'timing/model': 1.2191901900805533, 'learning_rate/action_model': 9.974227897558876e-05, 'learning_rate/action_query_proj': 9.974227897558876e-05, 'learning_rate/base': 2.493556974389719e-05, 'epoch': 0.03}) +[2026-06-29 15:26:44,587][starVLA.training.train_starvla][INFO] - Step 67, Loss: {'train/loss': 0.36329662799835205, 'train/grad_norm_pre_clip': 4.884199142456055, 'timing/data': 0.00025184499099850655, 'timing/model': 1.2192063331604004, 'learning_rate/action_model': 9.973441734728717e-05, 'learning_rate/action_query_proj': 9.973441734728717e-05, 'learning_rate/base': 2.4933604336821793e-05, 'epoch': 0.03}) +[2026-06-29 15:27:16,938][starVLA.training.train_starvla][INFO] - Step 68, Loss: {'train/loss': 0.47450053691864014, 'train/grad_norm_pre_clip': 2.679943799972534, 'timing/data': 0.00027086096815764904, 'timing/model': 1.2197146220132709, 'learning_rate/action_model': 9.972643793905595e-05, 'learning_rate/action_query_proj': 9.972643793905595e-05, 'learning_rate/base': 2.4931609484763988e-05, 'epoch': 0.03}) +[2026-06-29 15:27:49,314][starVLA.training.train_starvla][INFO] - Step 69, Loss: {'train/loss': 0.5131572484970093, 'train/grad_norm_pre_clip': 7.703522205352783, 'timing/data': 0.00018980796448886395, 'timing/model': 1.221724278992042, 'learning_rate/action_model': 9.971834077058345e-05, 'learning_rate/action_query_proj': 9.971834077058345e-05, 'learning_rate/base': 2.492958519264586e-05, 'epoch': 0.03}) +[2026-06-29 15:28:21,637][starVLA.training.train_starvla][INFO] - Step 70, Loss: {'train/loss': 0.1451525241136551, 'train/grad_norm_pre_clip': 6.961053848266602, 'timing/data': 0.0002012199256569147, 'timing/model': 1.2193651429843158, 'learning_rate/action_model': 9.971012586184863e-05, 'learning_rate/action_query_proj': 9.971012586184863e-05, 'learning_rate/base': 2.4927531465462158e-05, 'epoch': 0.03}) +[2026-06-29 15:28:54,011][starVLA.training.train_starvla][INFO] - Step 71, Loss: {'train/loss': 0.1070806011557579, 'train/grad_norm_pre_clip': 5.729844570159912, 'timing/data': 0.00022422289475798607, 'timing/model': 1.2199541539885104, 'learning_rate/action_model': 9.970179323312097e-05, 'learning_rate/action_query_proj': 9.970179323312097e-05, 'learning_rate/base': 2.492544830828024e-05, 'epoch': 0.03}) +[2026-06-29 15:29:26,360][starVLA.training.train_starvla][INFO] - Step 72, Loss: {'train/loss': 0.12900398671627045, 'train/grad_norm_pre_clip': 1.7418919801712036, 'timing/data': 0.00022037513554096222, 'timing/model': 1.2206721149850637, 'learning_rate/action_model': 9.96933429049604e-05, 'learning_rate/action_query_proj': 9.96933429049604e-05, 'learning_rate/base': 2.49233357262401e-05, 'epoch': 0.03}) +[2026-06-29 15:29:58,722][starVLA.training.train_starvla][INFO] - Step 73, Loss: {'train/loss': 0.4559861123561859, 'train/grad_norm_pre_clip': 1.964740514755249, 'timing/data': 0.00020050792954862118, 'timing/model': 1.2192897491622716, 'learning_rate/action_model': 9.968477489821726e-05, 'learning_rate/action_query_proj': 9.968477489821726e-05, 'learning_rate/base': 2.4921193724554316e-05, 'epoch': 0.03}) +[2026-06-29 15:30:31,058][starVLA.training.train_starvla][INFO] - Step 74, Loss: {'train/loss': 0.43443357944488525, 'train/grad_norm_pre_clip': 8.648050308227539, 'timing/data': 0.00019358494319021702, 'timing/model': 1.2194712099153548, 'learning_rate/action_model': 9.96760892340323e-05, 'learning_rate/action_query_proj': 9.96760892340323e-05, 'learning_rate/base': 2.4919022308508075e-05, 'epoch': 0.03}) +[2026-06-29 15:31:03,447][starVLA.training.train_starvla][INFO] - Step 75, Loss: {'train/loss': 1.0838019847869873, 'train/grad_norm_pre_clip': 8.55343246459961, 'timing/data': 0.00019402592442929745, 'timing/model': 1.2239992620889097, 'learning_rate/action_model': 9.966728593383647e-05, 'learning_rate/action_query_proj': 9.966728593383647e-05, 'learning_rate/base': 2.4916821483459117e-05, 'epoch': 0.03}) +[2026-06-29 15:31:35,787][starVLA.training.train_starvla][INFO] - Step 76, Loss: {'train/loss': 0.16659533977508545, 'train/grad_norm_pre_clip': 8.836591720581055, 'timing/data': 0.00029096798971295357, 'timing/model': 1.22157659009099, 'learning_rate/action_model': 9.965836501935105e-05, 'learning_rate/action_query_proj': 9.965836501935105e-05, 'learning_rate/base': 2.4914591254837762e-05, 'epoch': 0.03}) +[2026-06-29 15:32:08,102][starVLA.training.train_starvla][INFO] - Step 77, Loss: {'train/loss': 0.0322725884616375, 'train/grad_norm_pre_clip': 3.497767210006714, 'timing/data': 0.00031642685644328594, 'timing/model': 1.2207293550018221, 'learning_rate/action_model': 9.964932651258754e-05, 'learning_rate/action_query_proj': 9.964932651258754e-05, 'learning_rate/base': 2.4912331628146884e-05, 'epoch': 0.03}) +[2026-06-29 15:32:40,446][starVLA.training.train_starvla][INFO] - Step 78, Loss: {'train/loss': 0.44233882427215576, 'train/grad_norm_pre_clip': 5.0987868309021, 'timing/data': 0.0003877920098602772, 'timing/model': 1.2199378779623657, 'learning_rate/action_model': 9.964017043584755e-05, 'learning_rate/action_query_proj': 9.964017043584755e-05, 'learning_rate/base': 2.4910042608961888e-05, 'epoch': 0.03}) +[2026-06-29 15:33:12,797][starVLA.training.train_starvla][INFO] - Step 79, Loss: {'train/loss': 0.16738861799240112, 'train/grad_norm_pre_clip': 2.7671875953674316, 'timing/data': 0.00032772799022495747, 'timing/model': 1.2209650750737637, 'learning_rate/action_model': 9.963089681172276e-05, 'learning_rate/action_query_proj': 9.963089681172276e-05, 'learning_rate/base': 2.490772420293069e-05, 'epoch': 0.03}) +[2026-06-29 15:33:45,151][starVLA.training.train_starvla][INFO] - Step 80, Loss: {'train/loss': 0.12893074750900269, 'train/grad_norm_pre_clip': 9.736734390258789, 'timing/data': 0.0002522459253668785, 'timing/model': 1.2201838551554829, 'learning_rate/action_model': 9.962150566309495e-05, 'learning_rate/action_query_proj': 9.962150566309495e-05, 'learning_rate/base': 2.4905376415773737e-05, 'epoch': 0.03}) +[2026-06-29 15:34:17,481][starVLA.training.train_starvla][INFO] - Step 81, Loss: {'train/loss': 0.023156888782978058, 'train/grad_norm_pre_clip': 12.184470176696777, 'timing/data': 0.00025569205172359943, 'timing/model': 1.2207214660011232, 'learning_rate/action_model': 9.961199701313582e-05, 'learning_rate/action_query_proj': 9.961199701313582e-05, 'learning_rate/base': 2.4902999253283955e-05, 'epoch': 0.03}) +[2026-06-29 15:34:49,827][starVLA.training.train_starvla][INFO] - Step 82, Loss: {'train/loss': 0.6374895572662354, 'train/grad_norm_pre_clip': 13.79708480834961, 'timing/data': 0.0002056469675153494, 'timing/model': 1.2187602079939097, 'learning_rate/action_model': 9.960237088530703e-05, 'learning_rate/action_query_proj': 9.960237088530703e-05, 'learning_rate/base': 2.4900592721326758e-05, 'epoch': 0.04}) +[2026-06-29 15:35:22,169][starVLA.training.train_starvla][INFO] - Step 83, Loss: {'train/loss': 0.015138546004891396, 'train/grad_norm_pre_clip': 10.863615989685059, 'timing/data': 0.00028491695411503315, 'timing/model': 1.219718427862972, 'learning_rate/action_model': 9.959262730336007e-05, 'learning_rate/action_query_proj': 9.959262730336007e-05, 'learning_rate/base': 2.4898156825840018e-05, 'epoch': 0.04}) +[2026-06-29 15:35:54,480][starVLA.training.train_starvla][INFO] - Step 84, Loss: {'train/loss': 0.29275602102279663, 'train/grad_norm_pre_clip': 9.913618087768555, 'timing/data': 0.00019789300858974457, 'timing/model': 1.2194567320402712, 'learning_rate/action_model': 9.958276629133632e-05, 'learning_rate/action_query_proj': 9.958276629133632e-05, 'learning_rate/base': 2.489569157283408e-05, 'epoch': 0.04}) +[2026-06-29 15:36:26,869][starVLA.training.train_starvla][INFO] - Step 85, Loss: {'train/loss': 0.20316018164157867, 'train/grad_norm_pre_clip': 5.150519847869873, 'timing/data': 0.00022201798856258392, 'timing/model': 1.2193994119297713, 'learning_rate/action_model': 9.95727878735668e-05, 'learning_rate/action_query_proj': 9.95727878735668e-05, 'learning_rate/base': 2.48931969683917e-05, 'epoch': 0.04}) +[2026-06-29 15:36:59,204][starVLA.training.train_starvla][INFO] - Step 86, Loss: {'train/loss': 0.1944708228111267, 'train/grad_norm_pre_clip': 6.495004177093506, 'timing/data': 0.00028529809787869453, 'timing/model': 1.2209722299594432, 'learning_rate/action_model': 9.956269207467226e-05, 'learning_rate/action_query_proj': 9.956269207467226e-05, 'learning_rate/base': 2.4890673018668066e-05, 'epoch': 0.04}) +[2026-06-29 15:37:31,580][starVLA.training.train_starvla][INFO] - Step 87, Loss: {'train/loss': 0.15519998967647552, 'train/grad_norm_pre_clip': 4.081754684448242, 'timing/data': 0.00018946710042655468, 'timing/model': 1.2233171740081161, 'learning_rate/action_model': 9.955247891956309e-05, 'learning_rate/action_query_proj': 9.955247891956309e-05, 'learning_rate/base': 2.4888119729890773e-05, 'epoch': 0.04}) +[2026-06-29 15:38:03,882][starVLA.training.train_starvla][INFO] - Step 88, Loss: {'train/loss': 0.1860015094280243, 'train/grad_norm_pre_clip': 1.690287470817566, 'timing/data': 0.00030580698512494564, 'timing/model': 1.2225395939312875, 'learning_rate/action_model': 9.954214843343925e-05, 'learning_rate/action_query_proj': 9.954214843343925e-05, 'learning_rate/base': 2.4885537108359812e-05, 'epoch': 0.04}) +[2026-06-29 15:38:36,216][starVLA.training.train_starvla][INFO] - Step 89, Loss: {'train/loss': 0.5869157910346985, 'train/grad_norm_pre_clip': 6.372660160064697, 'timing/data': 0.00020340410992503166, 'timing/model': 1.2212566731031984, 'learning_rate/action_model': 9.95317006417902e-05, 'learning_rate/action_query_proj': 9.95317006417902e-05, 'learning_rate/base': 2.488292516044755e-05, 'epoch': 0.04}) +[2026-06-29 15:39:08,584][starVLA.training.train_starvla][INFO] - Step 90, Loss: {'train/loss': 0.04323694482445717, 'train/grad_norm_pre_clip': 6.516275882720947, 'timing/data': 0.0002106169704347849, 'timing/model': 1.2232451629824936, 'learning_rate/action_model': 9.952113557039477e-05, 'learning_rate/action_query_proj': 9.952113557039477e-05, 'learning_rate/base': 2.488028389259869e-05, 'epoch': 0.04}) +[2026-06-29 15:39:40,904][starVLA.training.train_starvla][INFO] - Step 91, Loss: {'train/loss': 0.09887474775314331, 'train/grad_norm_pre_clip': 4.6720123291015625, 'timing/data': 0.00019904505461454391, 'timing/model': 1.2220121589489281, 'learning_rate/action_model': 9.951045324532125e-05, 'learning_rate/action_query_proj': 9.951045324532125e-05, 'learning_rate/base': 2.4877613311330313e-05, 'epoch': 0.04}) +[2026-06-29 15:40:13,248][starVLA.training.train_starvla][INFO] - Step 92, Loss: {'train/loss': 0.061814796179533005, 'train/grad_norm_pre_clip': 4.854802131652832, 'timing/data': 0.00019441707991063595, 'timing/model': 1.2211911650374532, 'learning_rate/action_model': 9.949965369292724e-05, 'learning_rate/action_query_proj': 9.949965369292724e-05, 'learning_rate/base': 2.487491342323181e-05, 'epoch': 0.04}) +[2026-06-29 15:40:45,586][starVLA.training.train_starvla][INFO] - Step 93, Loss: {'train/loss': 0.14792948961257935, 'train/grad_norm_pre_clip': 2.712627649307251, 'timing/data': 0.00024637486785650253, 'timing/model': 1.221317619085312, 'learning_rate/action_model': 9.948873693985955e-05, 'learning_rate/action_query_proj': 9.948873693985955e-05, 'learning_rate/base': 2.4872184234964888e-05, 'epoch': 0.04}) +[2026-06-29 15:41:17,931][starVLA.training.train_starvla][INFO] - Step 94, Loss: {'train/loss': 0.08654066175222397, 'train/grad_norm_pre_clip': 3.614025354385376, 'timing/data': 0.0004133698530495167, 'timing/model': 1.2209428590722382, 'learning_rate/action_model': 9.947770301305417e-05, 'learning_rate/action_query_proj': 9.947770301305417e-05, 'learning_rate/base': 2.486942575326354e-05, 'epoch': 0.04}) +[2026-06-29 15:41:50,424][starVLA.training.train_starvla][INFO] - Step 95, Loss: {'train/loss': 0.21256761252880096, 'train/grad_norm_pre_clip': 4.39247989654541, 'timing/data': 0.00019607017748057842, 'timing/model': 1.3663403380196542, 'learning_rate/action_model': 9.946655193973622e-05, 'learning_rate/action_query_proj': 9.946655193973622e-05, 'learning_rate/base': 2.4866637984934056e-05, 'epoch': 0.04}) +[2026-06-29 15:42:22,775][starVLA.training.train_starvla][INFO] - Step 96, Loss: {'train/loss': 0.003702008631080389, 'train/grad_norm_pre_clip': 7.761146545410156, 'timing/data': 0.00041443086229264736, 'timing/model': 1.21977564599365, 'learning_rate/action_model': 9.945528374741986e-05, 'learning_rate/action_query_proj': 9.945528374741986e-05, 'learning_rate/base': 2.4863820936854966e-05, 'epoch': 0.04}) +[2026-06-29 15:42:55,115][starVLA.training.train_starvla][INFO] - Step 97, Loss: {'train/loss': 0.2008773684501648, 'train/grad_norm_pre_clip': 12.381869316101074, 'timing/data': 0.00019586901180446148, 'timing/model': 1.2219104720279574, 'learning_rate/action_model': 9.944389846390828e-05, 'learning_rate/action_query_proj': 9.944389846390828e-05, 'learning_rate/base': 2.486097461597707e-05, 'epoch': 0.04}) +[2026-06-29 15:43:27,407][starVLA.training.train_starvla][INFO] - Step 98, Loss: {'train/loss': 0.055421002209186554, 'train/grad_norm_pre_clip': 7.727570056915283, 'timing/data': 0.00024857884272933006, 'timing/model': 1.2154818740673363, 'learning_rate/action_model': 9.943239611729349e-05, 'learning_rate/action_query_proj': 9.943239611729349e-05, 'learning_rate/base': 2.485809902932337e-05, 'epoch': 0.04}) +[2026-06-29 15:43:59,731][starVLA.training.train_starvla][INFO] - Step 99, Loss: {'train/loss': 0.19996054470539093, 'train/grad_norm_pre_clip': 3.1384048461914062, 'timing/data': 0.00020321295596659184, 'timing/model': 1.227809488074854, 'learning_rate/action_model': 9.942077673595636e-05, 'learning_rate/action_query_proj': 9.942077673595636e-05, 'learning_rate/base': 2.485519418398909e-05, 'epoch': 0.04}) +[2026-06-29 15:44:32,095][starVLA.training.train_starvla][INFO] - Step 100, Loss: {'train/loss': 0.16511967778205872, 'train/grad_norm_pre_clip': 2.3618340492248535, 'timing/data': 0.0002364658284932375, 'timing/model': 1.2204947208520025, 'learning_rate/action_model': 9.940904034856662e-05, 'learning_rate/action_query_proj': 9.940904034856662e-05, 'learning_rate/base': 2.4852260087141655e-05, 'epoch': 0.04}) +[2026-06-29 15:45:04,467][starVLA.training.train_starvla][INFO] - Step 101, Loss: {'train/loss': 0.24561354517936707, 'train/grad_norm_pre_clip': 1.696498990058899, 'timing/data': 0.00025568180717527866, 'timing/model': 1.2183952888008207, 'learning_rate/action_model': 9.93971869840826e-05, 'learning_rate/action_query_proj': 9.93971869840826e-05, 'learning_rate/base': 2.484929674602065e-05, 'epoch': 0.04}) +[2026-06-29 15:45:36,796][starVLA.training.train_starvla][INFO] - Step 102, Loss: {'train/loss': 0.061531055718660355, 'train/grad_norm_pre_clip': 5.81192684173584, 'timing/data': 0.00027325493283569813, 'timing/model': 1.2219898619223386, 'learning_rate/action_model': 9.938521667175133e-05, 'learning_rate/action_query_proj': 9.938521667175133e-05, 'learning_rate/base': 2.484630416793783e-05, 'epoch': 0.04}) +[2026-06-29 15:46:09,126][starVLA.training.train_starvla][INFO] - Step 103, Loss: {'train/loss': 1.2260348796844482, 'train/grad_norm_pre_clip': 10.662893295288086, 'timing/data': 0.0003656789194792509, 'timing/model': 1.218478088034317, 'learning_rate/action_model': 9.937312944110834e-05, 'learning_rate/action_query_proj': 9.937312944110834e-05, 'learning_rate/base': 2.4843282360277086e-05, 'epoch': 0.04}) +[2026-06-29 15:46:41,487][starVLA.training.train_starvla][INFO] - Step 104, Loss: {'train/loss': 0.23452092707157135, 'train/grad_norm_pre_clip': 6.101410865783691, 'timing/data': 0.00028841407038271427, 'timing/model': 1.2170655231457204, 'learning_rate/action_model': 9.936092532197766e-05, 'learning_rate/action_query_proj': 9.936092532197766e-05, 'learning_rate/base': 2.4840231330494415e-05, 'epoch': 0.04}) +[2026-06-29 15:47:13,839][starVLA.training.train_starvla][INFO] - Step 105, Loss: {'train/loss': 0.1293104737997055, 'train/grad_norm_pre_clip': 4.672348976135254, 'timing/data': 0.00027620093896985054, 'timing/model': 1.2211906779557467, 'learning_rate/action_model': 9.934860434447179e-05, 'learning_rate/action_query_proj': 9.934860434447179e-05, 'learning_rate/base': 2.4837151086117947e-05, 'epoch': 0.05}) +[2026-06-29 15:47:46,182][starVLA.training.train_starvla][INFO] - Step 106, Loss: {'train/loss': 0.043496061116456985, 'train/grad_norm_pre_clip': 5.668693542480469, 'timing/data': 0.00018691201694309711, 'timing/model': 1.2199560089502484, 'learning_rate/action_model': 9.933616653899147e-05, 'learning_rate/action_query_proj': 9.933616653899147e-05, 'learning_rate/base': 2.4834041634747868e-05, 'epoch': 0.05}) +[2026-06-29 15:48:18,490][starVLA.training.train_starvla][INFO] - Step 107, Loss: {'train/loss': 0.018564235419034958, 'train/grad_norm_pre_clip': 1.580911636352539, 'timing/data': 0.0003630740102380514, 'timing/model': 1.2206993610598147, 'learning_rate/action_model': 9.932361193622577e-05, 'learning_rate/action_query_proj': 9.932361193622577e-05, 'learning_rate/base': 2.4830902984056442e-05, 'epoch': 0.05}) +[2026-06-29 15:48:50,779][starVLA.training.train_starvla][INFO] - Step 108, Loss: {'train/loss': 0.04092200845479965, 'train/grad_norm_pre_clip': 4.277093410491943, 'timing/data': 0.00024367892183363438, 'timing/model': 1.218195830937475, 'learning_rate/action_model': 9.931094056715191e-05, 'learning_rate/action_query_proj': 9.931094056715191e-05, 'learning_rate/base': 2.4827735141787978e-05, 'epoch': 0.05}) +[2026-06-29 15:49:23,087][starVLA.training.train_starvla][INFO] - Step 109, Loss: {'train/loss': 0.07288160175085068, 'train/grad_norm_pre_clip': 9.21072769165039, 'timing/data': 0.00024420092813670635, 'timing/model': 1.223241083091125, 'learning_rate/action_model': 9.929815246303525e-05, 'learning_rate/action_query_proj': 9.929815246303525e-05, 'learning_rate/base': 2.4824538115758812e-05, 'epoch': 0.05}) +[2026-06-29 15:49:55,421][starVLA.training.train_starvla][INFO] - Step 110, Loss: {'train/loss': 0.19510959088802338, 'train/grad_norm_pre_clip': 10.439056396484375, 'timing/data': 0.0002859989181160927, 'timing/model': 1.2190290950238705, 'learning_rate/action_model': 9.928524765542915e-05, 'learning_rate/action_query_proj': 9.928524765542915e-05, 'learning_rate/base': 2.4821311913857286e-05, 'epoch': 0.05}) +[2026-06-29 15:50:27,775][starVLA.training.train_starvla][INFO] - Step 111, Loss: {'train/loss': 0.25591719150543213, 'train/grad_norm_pre_clip': 2.483492136001587, 'timing/data': 0.00019492791034281254, 'timing/model': 1.22146717319265, 'learning_rate/action_model': 9.927222617617495e-05, 'learning_rate/action_query_proj': 9.927222617617495e-05, 'learning_rate/base': 2.4818056544043738e-05, 'epoch': 0.05}) +[2026-06-29 15:51:00,136][starVLA.training.train_starvla][INFO] - Step 112, Loss: {'train/loss': 0.005724628455936909, 'train/grad_norm_pre_clip': 5.6036458015441895, 'timing/data': 0.0001986050046980381, 'timing/model': 1.2187463559675962, 'learning_rate/action_model': 9.925908805740186e-05, 'learning_rate/action_query_proj': 9.925908805740186e-05, 'learning_rate/base': 2.4814772014350464e-05, 'epoch': 0.05}) +[2026-06-29 15:51:32,475][starVLA.training.train_starvla][INFO] - Step 113, Loss: {'train/loss': 0.10564734786748886, 'train/grad_norm_pre_clip': 8.000781059265137, 'timing/data': 0.00019326596520841122, 'timing/model': 1.22226003697142, 'learning_rate/action_model': 9.924583333152687e-05, 'learning_rate/action_query_proj': 9.924583333152687e-05, 'learning_rate/base': 2.481145833288172e-05, 'epoch': 0.05}) +[2026-06-29 15:52:04,820][starVLA.training.train_starvla][INFO] - Step 114, Loss: {'train/loss': 0.16601711511611938, 'train/grad_norm_pre_clip': 1.782273530960083, 'timing/data': 0.0001913420855998993, 'timing/model': 1.2302574519999325, 'learning_rate/action_model': 9.923246203125471e-05, 'learning_rate/action_query_proj': 9.923246203125471e-05, 'learning_rate/base': 2.480811550781368e-05, 'epoch': 0.05}) +[2026-06-29 15:52:37,187][starVLA.training.train_starvla][INFO] - Step 115, Loss: {'train/loss': 0.33997854590415955, 'train/grad_norm_pre_clip': 6.95366907119751, 'timing/data': 0.00019208411686122417, 'timing/model': 1.2218146859668195, 'learning_rate/action_model': 9.921897418957773e-05, 'learning_rate/action_query_proj': 9.921897418957773e-05, 'learning_rate/base': 2.4804743547394433e-05, 'epoch': 0.05}) +[2026-06-29 15:53:09,508][starVLA.training.train_starvla][INFO] - Step 116, Loss: {'train/loss': 0.049494899809360504, 'train/grad_norm_pre_clip': 7.433032035827637, 'timing/data': 0.0001988760195672512, 'timing/model': 1.2188844550400972, 'learning_rate/action_model': 9.920536983977583e-05, 'learning_rate/action_query_proj': 9.920536983977583e-05, 'learning_rate/base': 2.4801342459943956e-05, 'epoch': 0.05}) +[2026-06-29 15:53:41,820][starVLA.training.train_starvla][INFO] - Step 117, Loss: {'train/loss': 0.003677364205941558, 'train/grad_norm_pre_clip': 7.483353614807129, 'timing/data': 0.0002468570601195097, 'timing/model': 1.2203680099919438, 'learning_rate/action_model': 9.91916490154164e-05, 'learning_rate/action_query_proj': 9.91916490154164e-05, 'learning_rate/base': 2.47979122538541e-05, 'epoch': 0.05}) +[2026-06-29 15:54:14,166][starVLA.training.train_starvla][INFO] - Step 118, Loss: {'train/loss': 0.08101330697536469, 'train/grad_norm_pre_clip': 3.073505401611328, 'timing/data': 0.0002028129529207945, 'timing/model': 1.2214838720392436, 'learning_rate/action_model': 9.91778117503542e-05, 'learning_rate/action_query_proj': 9.91778117503542e-05, 'learning_rate/base': 2.479445293758855e-05, 'epoch': 0.05}) +[2026-06-29 15:54:46,551][starVLA.training.train_starvla][INFO] - Step 119, Loss: {'train/loss': 0.3777558207511902, 'train/grad_norm_pre_clip': 3.0587213039398193, 'timing/data': 0.0001895881723612547, 'timing/model': 1.222060828935355, 'learning_rate/action_model': 9.916385807873133e-05, 'learning_rate/action_query_proj': 9.916385807873133e-05, 'learning_rate/base': 2.4790964519682833e-05, 'epoch': 0.05}) +[2026-06-29 15:55:18,911][starVLA.training.train_starvla][INFO] - Step 120, Loss: {'train/loss': 0.09003492444753647, 'train/grad_norm_pre_clip': 2.0338823795318604, 'timing/data': 0.0002466561272740364, 'timing/model': 1.221307174069807, 'learning_rate/action_model': 9.914978803497707e-05, 'learning_rate/action_query_proj': 9.914978803497707e-05, 'learning_rate/base': 2.4787447008744268e-05, 'epoch': 0.05}) +[2026-06-29 15:55:51,288][starVLA.training.train_starvla][INFO] - Step 121, Loss: {'train/loss': 0.029218383133411407, 'train/grad_norm_pre_clip': 11.75892448425293, 'timing/data': 0.00019489903934299946, 'timing/model': 1.2213908601552248, 'learning_rate/action_model': 9.913560165380783e-05, 'learning_rate/action_query_proj': 9.913560165380783e-05, 'learning_rate/base': 2.478390041345196e-05, 'epoch': 0.05}) +[2026-06-29 15:56:23,633][starVLA.training.train_starvla][INFO] - Step 122, Loss: {'train/loss': 0.10729645937681198, 'train/grad_norm_pre_clip': 11.897294998168945, 'timing/data': 0.00019351579248905182, 'timing/model': 1.2208393979817629, 'learning_rate/action_model': 9.912129897022714e-05, 'learning_rate/action_query_proj': 9.912129897022714e-05, 'learning_rate/base': 2.4780324742556786e-05, 'epoch': 0.05}) +[2026-06-29 15:56:55,928][starVLA.training.train_starvla][INFO] - Step 123, Loss: {'train/loss': 0.33768144249916077, 'train/grad_norm_pre_clip': 8.251900672912598, 'timing/data': 0.00019056000746786594, 'timing/model': 1.2192573968786746, 'learning_rate/action_model': 9.910688001952544e-05, 'learning_rate/action_query_proj': 9.910688001952544e-05, 'learning_rate/base': 2.477672000488136e-05, 'epoch': 0.05}) +[2026-06-29 15:57:28,218][starVLA.training.train_starvla][INFO] - Step 124, Loss: {'train/loss': 0.015817692503333092, 'train/grad_norm_pre_clip': 4.004604816436768, 'timing/data': 0.0002879940439015627, 'timing/model': 1.2232576289679855, 'learning_rate/action_model': 9.909234483728003e-05, 'learning_rate/action_query_proj': 9.909234483728003e-05, 'learning_rate/base': 2.477308620932001e-05, 'epoch': 0.05}) +[2026-06-29 15:58:00,571][starVLA.training.train_starvla][INFO] - Step 125, Loss: {'train/loss': 0.028741028159856796, 'train/grad_norm_pre_clip': 2.8617262840270996, 'timing/data': 0.00028507900424301624, 'timing/model': 1.2178597741294652, 'learning_rate/action_model': 9.907769345935507e-05, 'learning_rate/action_query_proj': 9.907769345935507e-05, 'learning_rate/base': 2.4769423364838768e-05, 'epoch': 0.05}) +[2026-06-29 15:58:32,900][starVLA.training.train_starvla][INFO] - Step 126, Loss: {'train/loss': 0.01568889245390892, 'train/grad_norm_pre_clip': 4.311301231384277, 'timing/data': 0.0003975299187004566, 'timing/model': 1.2218047929927707, 'learning_rate/action_model': 9.906292592190133e-05, 'learning_rate/action_query_proj': 9.906292592190133e-05, 'learning_rate/base': 2.4765731480475333e-05, 'epoch': 0.05}) +[2026-06-29 15:59:05,214][starVLA.training.train_starvla][INFO] - Step 127, Loss: {'train/loss': 0.03681352362036705, 'train/grad_norm_pre_clip': 9.291507720947266, 'timing/data': 0.00020111002959311008, 'timing/model': 1.2173306629993021, 'learning_rate/action_model': 9.904804226135628e-05, 'learning_rate/action_query_proj': 9.904804226135628e-05, 'learning_rate/base': 2.476201056533907e-05, 'epoch': 0.05}) +[2026-06-29 15:59:37,549][starVLA.training.train_starvla][INFO] - Step 128, Loss: {'train/loss': 0.45072343945503235, 'train/grad_norm_pre_clip': 6.2951483726501465, 'timing/data': 0.0002482789568603039, 'timing/model': 1.2197877389844507, 'learning_rate/action_model': 9.903304251444385e-05, 'learning_rate/action_query_proj': 9.903304251444385e-05, 'learning_rate/base': 2.4758260628610964e-05, 'epoch': 0.06}) +[2026-06-29 16:00:09,887][starVLA.training.train_starvla][INFO] - Step 129, Loss: {'train/loss': 0.005812902934849262, 'train/grad_norm_pre_clip': 1.9225122928619385, 'timing/data': 0.00025665503926575184, 'timing/model': 1.2203050530515611, 'learning_rate/action_model': 9.901792671817444e-05, 'learning_rate/action_query_proj': 9.901792671817444e-05, 'learning_rate/base': 2.475448167954361e-05, 'epoch': 0.06}) +[2026-06-29 16:00:42,207][starVLA.training.train_starvla][INFO] - Step 130, Loss: {'train/loss': 0.1268664300441742, 'train/grad_norm_pre_clip': 3.3909707069396973, 'timing/data': 0.0002075820229947567, 'timing/model': 1.221805094042793, 'learning_rate/action_model': 9.900269490984477e-05, 'learning_rate/action_query_proj': 9.900269490984477e-05, 'learning_rate/base': 2.4750673727461192e-05, 'epoch': 0.06}) +[2026-06-29 16:01:14,521][starVLA.training.train_starvla][INFO] - Step 131, Loss: {'train/loss': 0.013368134386837482, 'train/grad_norm_pre_clip': 1.1911510229110718, 'timing/data': 0.0001982138492166996, 'timing/model': 1.2232038339134306, 'learning_rate/action_model': 9.89873471270378e-05, 'learning_rate/action_query_proj': 9.89873471270378e-05, 'learning_rate/base': 2.474683678175945e-05, 'epoch': 0.06}) +[2026-06-29 16:01:46,855][starVLA.training.train_starvla][INFO] - Step 132, Loss: {'train/loss': 0.04863736405968666, 'train/grad_norm_pre_clip': 7.114891052246094, 'timing/data': 0.0002450819592922926, 'timing/model': 1.2220241480972618, 'learning_rate/action_model': 9.897188340762267e-05, 'learning_rate/action_query_proj': 9.897188340762267e-05, 'learning_rate/base': 2.4742970851905668e-05, 'epoch': 0.06}) +[2026-06-29 16:02:19,147][starVLA.training.train_starvla][INFO] - Step 133, Loss: {'train/loss': 0.20665378868579865, 'train/grad_norm_pre_clip': 5.440195083618164, 'timing/data': 0.0002447820734232664, 'timing/model': 1.2225699978880584, 'learning_rate/action_model': 9.895630378975456e-05, 'learning_rate/action_query_proj': 9.895630378975456e-05, 'learning_rate/base': 2.473907594743864e-05, 'epoch': 0.06}) +[2026-06-29 16:02:51,481][starVLA.training.train_starvla][INFO] - Step 134, Loss: {'train/loss': 0.025904662907123566, 'train/grad_norm_pre_clip': 2.0608203411102295, 'timing/data': 0.0001963099930435419, 'timing/model': 1.211042997892946, 'learning_rate/action_model': 9.894060831187465e-05, 'learning_rate/action_query_proj': 9.894060831187465e-05, 'learning_rate/base': 2.4735152077968663e-05, 'epoch': 0.06}) +[2026-06-29 16:03:23,935][starVLA.training.train_starvla][INFO] - Step 135, Loss: {'train/loss': 0.0010485145030543208, 'train/grad_norm_pre_clip': 5.182494163513184, 'timing/data': 0.00045241392217576504, 'timing/model': 1.2164303560275584, 'learning_rate/action_model': 9.892479701270997e-05, 'learning_rate/action_query_proj': 9.892479701270997e-05, 'learning_rate/base': 2.4731199253177493e-05, 'epoch': 0.06}) +[2026-06-29 16:03:56,236][starVLA.training.train_starvla][INFO] - Step 136, Loss: {'train/loss': 0.059707868844270706, 'train/grad_norm_pre_clip': 1.3522640466690063, 'timing/data': 0.00020080897957086563, 'timing/model': 1.2189480238594115, 'learning_rate/action_model': 9.89088699312733e-05, 'learning_rate/action_query_proj': 9.89088699312733e-05, 'learning_rate/base': 2.4727217482818323e-05, 'epoch': 0.06}) +[2026-06-29 16:04:28,577][starVLA.training.train_starvla][INFO] - Step 137, Loss: {'train/loss': 0.1934383660554886, 'train/grad_norm_pre_clip': 2.5172643661499023, 'timing/data': 0.0001977530773729086, 'timing/model': 1.2228218079544604, 'learning_rate/action_model': 9.889282710686314e-05, 'learning_rate/action_query_proj': 9.889282710686314e-05, 'learning_rate/base': 2.4723206776715786e-05, 'epoch': 0.06}) +[2026-06-29 16:05:00,899][starVLA.training.train_starvla][INFO] - Step 138, Loss: {'train/loss': 0.08822797238826752, 'train/grad_norm_pre_clip': 4.109055519104004, 'timing/data': 0.00025383918546140194, 'timing/model': 1.2216740529984236, 'learning_rate/action_model': 9.887666857906358e-05, 'learning_rate/action_query_proj': 9.887666857906358e-05, 'learning_rate/base': 2.4719167144765895e-05, 'epoch': 0.06}) +[2026-06-29 16:05:33,244][starVLA.training.train_starvla][INFO] - Step 139, Loss: {'train/loss': 0.6730089783668518, 'train/grad_norm_pre_clip': 2.199846029281616, 'timing/data': 0.00029465602710843086, 'timing/model': 1.2219753691460937, 'learning_rate/action_model': 9.886039438774418e-05, 'learning_rate/action_query_proj': 9.886039438774418e-05, 'learning_rate/base': 2.4715098596936045e-05, 'epoch': 0.06}) +[2026-06-29 16:06:05,613][starVLA.training.train_starvla][INFO] - Step 140, Loss: {'train/loss': 0.0662417858839035, 'train/grad_norm_pre_clip': 3.6419756412506104, 'timing/data': 0.00023597478866577148, 'timing/model': 1.2221544510684907, 'learning_rate/action_model': 9.884400457305988e-05, 'learning_rate/action_query_proj': 9.884400457305988e-05, 'learning_rate/base': 2.471100114326497e-05, 'epoch': 0.06}) +[2026-06-29 16:06:38,149][starVLA.training.train_starvla][INFO] - Step 141, Loss: {'train/loss': 0.03874324634671211, 'train/grad_norm_pre_clip': 2.5247557163238525, 'timing/data': 0.0001920729409903288, 'timing/model': 1.2189646100159734, 'learning_rate/action_model': 9.882749917545091e-05, 'learning_rate/action_query_proj': 9.882749917545091e-05, 'learning_rate/base': 2.470687479386273e-05, 'epoch': 0.06}) +[2026-06-29 16:07:10,519][starVLA.training.train_starvla][INFO] - Step 142, Loss: {'train/loss': 0.14721593260765076, 'train/grad_norm_pre_clip': 3.2164814472198486, 'timing/data': 0.0002041049301624298, 'timing/model': 1.2213270128704607, 'learning_rate/action_model': 9.881087823564272e-05, 'learning_rate/action_query_proj': 9.881087823564272e-05, 'learning_rate/base': 2.470271955891068e-05, 'epoch': 0.06}) +[2026-06-29 16:07:42,859][starVLA.training.train_starvla][INFO] - Step 143, Loss: {'train/loss': 0.005918469280004501, 'train/grad_norm_pre_clip': 3.206326961517334, 'timing/data': 0.0004971488378942013, 'timing/model': 1.2187251751311123, 'learning_rate/action_model': 9.879414179464584e-05, 'learning_rate/action_query_proj': 9.879414179464584e-05, 'learning_rate/base': 2.469853544866146e-05, 'epoch': 0.06}) +[2026-06-29 16:08:15,181][starVLA.training.train_starvla][INFO] - Step 144, Loss: {'train/loss': 0.013988944701850414, 'train/grad_norm_pre_clip': 2.6548352241516113, 'timing/data': 0.00028981687501072884, 'timing/model': 1.2174964908044785, 'learning_rate/action_model': 9.877728989375571e-05, 'learning_rate/action_query_proj': 9.877728989375571e-05, 'learning_rate/base': 2.4694322473438927e-05, 'epoch': 0.06}) +[2026-06-29 16:08:47,520][starVLA.training.train_starvla][INFO] - Step 145, Loss: {'train/loss': 0.13909579813480377, 'train/grad_norm_pre_clip': 3.2483909130096436, 'timing/data': 0.000290286960080266, 'timing/model': 1.2109258740674704, 'learning_rate/action_model': 9.876032257455277e-05, 'learning_rate/action_query_proj': 9.876032257455277e-05, 'learning_rate/base': 2.4690080643638193e-05, 'epoch': 0.06}) +[2026-06-29 16:09:19,858][starVLA.training.train_starvla][INFO] - Step 146, Loss: {'train/loss': 0.13153508305549622, 'train/grad_norm_pre_clip': 1.6538763046264648, 'timing/data': 0.0002421368844807148, 'timing/model': 1.2196282581426203, 'learning_rate/action_model': 9.874323987890218e-05, 'learning_rate/action_query_proj': 9.874323987890218e-05, 'learning_rate/base': 2.4685809969725546e-05, 'epoch': 0.06}) +[2026-06-29 16:09:52,183][starVLA.training.train_starvla][INFO] - Step 147, Loss: {'train/loss': 0.13570564985275269, 'train/grad_norm_pre_clip': 4.577687740325928, 'timing/data': 0.0002308760304003954, 'timing/model': 1.2234058841131628, 'learning_rate/action_model': 9.87260418489538e-05, 'learning_rate/action_query_proj': 9.87260418489538e-05, 'learning_rate/base': 2.468151046223845e-05, 'epoch': 0.06}) +[2026-06-29 16:10:24,538][starVLA.training.train_starvla][INFO] - Step 148, Loss: {'train/loss': 0.628676176071167, 'train/grad_norm_pre_clip': 1.8788281679153442, 'timing/data': 0.00019853399135172367, 'timing/model': 1.2204048440326005, 'learning_rate/action_model': 9.870872852714209e-05, 'learning_rate/action_query_proj': 9.870872852714209e-05, 'learning_rate/base': 2.4677182131785522e-05, 'epoch': 0.06}) +[2026-06-29 16:10:56,855][starVLA.training.train_starvla][INFO] - Step 149, Loss: {'train/loss': 0.10792507976293564, 'train/grad_norm_pre_clip': 1.468865990638733, 'timing/data': 0.0002979820128530264, 'timing/model': 1.2192150838673115, 'learning_rate/action_model': 9.869129995618587e-05, 'learning_rate/action_query_proj': 9.869129995618587e-05, 'learning_rate/base': 2.467282498904647e-05, 'epoch': 0.06}) +[2026-06-29 16:11:29,234][starVLA.training.train_starvla][INFO] - Step 150, Loss: {'train/loss': 0.29105064272880554, 'train/grad_norm_pre_clip': 3.2116971015930176, 'timing/data': 0.0003304840065538883, 'timing/model': 1.2178434550296515, 'learning_rate/action_model': 9.867375617908848e-05, 'learning_rate/action_query_proj': 9.867375617908848e-05, 'learning_rate/base': 2.466843904477212e-05, 'epoch': 0.06}) +[2026-06-29 16:12:01,554][starVLA.training.train_starvla][INFO] - Step 151, Loss: {'train/loss': 0.206185445189476, 'train/grad_norm_pre_clip': 3.2817606925964355, 'timing/data': 0.00024586403742432594, 'timing/model': 1.2203837600536644, 'learning_rate/action_model': 9.86560972391374e-05, 'learning_rate/action_query_proj': 9.86560972391374e-05, 'learning_rate/base': 2.466402430978435e-05, 'epoch': 0.06}) +[2026-06-29 16:12:33,938][starVLA.training.train_starvla][INFO] - Step 152, Loss: {'train/loss': 0.02281814254820347, 'train/grad_norm_pre_clip': 3.9868533611297607, 'timing/data': 0.00019169109873473644, 'timing/model': 1.2229409881401807, 'learning_rate/action_model': 9.863832317990436e-05, 'learning_rate/action_query_proj': 9.863832317990436e-05, 'learning_rate/base': 2.465958079497609e-05, 'epoch': 0.07}) +[2026-06-29 16:13:06,309][starVLA.training.train_starvla][INFO] - Step 153, Loss: {'train/loss': 0.01344700250774622, 'train/grad_norm_pre_clip': 1.7812604904174805, 'timing/data': 0.0002730749547481537, 'timing/model': 1.224364381050691, 'learning_rate/action_model': 9.862043404524502e-05, 'learning_rate/action_query_proj': 9.862043404524502e-05, 'learning_rate/base': 2.4655108511311255e-05, 'epoch': 0.07}) +[2026-06-29 16:13:38,670][starVLA.training.train_starvla][INFO] - Step 154, Loss: {'train/loss': 0.004467342980206013, 'train/grad_norm_pre_clip': 3.0156240463256836, 'timing/data': 0.0002473169006407261, 'timing/model': 1.2220960529521108, 'learning_rate/action_model': 9.86024298792991e-05, 'learning_rate/action_query_proj': 9.86024298792991e-05, 'learning_rate/base': 2.4650607469824774e-05, 'epoch': 0.07}) +[2026-06-29 16:14:11,026][starVLA.training.train_starvla][INFO] - Step 155, Loss: {'train/loss': 0.3555920720100403, 'train/grad_norm_pre_clip': 3.1311893463134766, 'timing/data': 0.0003475961275398731, 'timing/model': 1.2191901439800858, 'learning_rate/action_model': 9.858431072649005e-05, 'learning_rate/action_query_proj': 9.858431072649005e-05, 'learning_rate/base': 2.464607768162251e-05, 'epoch': 0.07}) +[2026-06-29 16:14:43,347][starVLA.training.train_starvla][INFO] - Step 156, Loss: {'train/loss': 0.04271847382187843, 'train/grad_norm_pre_clip': 5.867161273956299, 'timing/data': 0.0002487390302121639, 'timing/model': 1.217501071980223, 'learning_rate/action_model': 9.856607663152508e-05, 'learning_rate/action_query_proj': 9.856607663152508e-05, 'learning_rate/base': 2.464151915788127e-05, 'epoch': 0.07}) +[2026-06-29 16:15:15,668][starVLA.training.train_starvla][INFO] - Step 157, Loss: {'train/loss': 0.05147796496748924, 'train/grad_norm_pre_clip': 9.547118186950684, 'timing/data': 0.0001885660458356142, 'timing/model': 1.2198076909407973, 'learning_rate/action_model': 9.854772763939504e-05, 'learning_rate/action_query_proj': 9.854772763939504e-05, 'learning_rate/base': 2.463693190984876e-05, 'epoch': 0.07}) +[2026-06-29 16:15:47,983][starVLA.training.train_starvla][INFO] - Step 158, Loss: {'train/loss': 0.10032401233911514, 'train/grad_norm_pre_clip': 2.7019336223602295, 'timing/data': 0.00020111002959311008, 'timing/model': 1.221019936958328, 'learning_rate/action_model': 9.852926379537421e-05, 'learning_rate/action_query_proj': 9.852926379537421e-05, 'learning_rate/base': 2.4632315948843552e-05, 'epoch': 0.07}) +[2026-06-29 16:16:20,328][starVLA.training.train_starvla][INFO] - Step 159, Loss: {'train/loss': 0.08266791701316833, 'train/grad_norm_pre_clip': 6.274393081665039, 'timing/data': 0.00020166998729109764, 'timing/model': 1.2202329928986728, 'learning_rate/action_model': 9.851068514502031e-05, 'learning_rate/action_query_proj': 9.851068514502031e-05, 'learning_rate/base': 2.4627671286255078e-05, 'epoch': 0.07}) +[2026-06-29 16:16:52,697][starVLA.training.train_starvla][INFO] - Step 160, Loss: {'train/loss': 0.22281408309936523, 'train/grad_norm_pre_clip': 3.751168966293335, 'timing/data': 0.00019324407912790775, 'timing/model': 1.220153836067766, 'learning_rate/action_model': 9.849199173417429e-05, 'learning_rate/action_query_proj': 9.849199173417429e-05, 'learning_rate/base': 2.4622997933543573e-05, 'epoch': 0.07}) +[2026-06-29 16:17:25,064][starVLA.training.train_starvla][INFO] - Step 161, Loss: {'train/loss': 0.26958364248275757, 'train/grad_norm_pre_clip': 5.091404914855957, 'timing/data': 0.0003268369473516941, 'timing/model': 1.2226104950532317, 'learning_rate/action_model': 9.847318360896032e-05, 'learning_rate/action_query_proj': 9.847318360896032e-05, 'learning_rate/base': 2.461829590224008e-05, 'epoch': 0.07}) +[2026-06-29 16:17:57,427][starVLA.training.train_starvla][INFO] - Step 162, Loss: {'train/loss': 0.34162387251853943, 'train/grad_norm_pre_clip': 5.698487281799316, 'timing/data': 0.00019815401174128056, 'timing/model': 1.2236856389790773, 'learning_rate/action_model': 9.845426081578554e-05, 'learning_rate/action_query_proj': 9.845426081578554e-05, 'learning_rate/base': 2.4613565203946386e-05, 'epoch': 0.07}) +[2026-06-29 16:18:29,761][starVLA.training.train_starvla][INFO] - Step 163, Loss: {'train/loss': 0.4190279245376587, 'train/grad_norm_pre_clip': 1.900206208229065, 'timing/data': 0.0003083120100200176, 'timing/model': 1.2235245420597494, 'learning_rate/action_model': 9.84352234013401e-05, 'learning_rate/action_query_proj': 9.84352234013401e-05, 'learning_rate/base': 2.4608805850335025e-05, 'epoch': 0.07}) +[2026-06-29 16:19:02,151][starVLA.training.train_starvla][INFO] - Step 164, Loss: {'train/loss': 0.27472206950187683, 'train/grad_norm_pre_clip': 1.3970543146133423, 'timing/data': 0.00039033591747283936, 'timing/model': 1.2207442270591855, 'learning_rate/action_model': 9.841607141259689e-05, 'learning_rate/action_query_proj': 9.841607141259689e-05, 'learning_rate/base': 2.4604017853149223e-05, 'epoch': 0.07}) +[2026-06-29 16:19:34,505][starVLA.training.train_starvla][INFO] - Step 165, Loss: {'train/loss': 0.06474093347787857, 'train/grad_norm_pre_clip': 2.2847890853881836, 'timing/data': 0.0001974820625036955, 'timing/model': 1.220502802869305, 'learning_rate/action_model': 9.839680489681155e-05, 'learning_rate/action_query_proj': 9.839680489681155e-05, 'learning_rate/base': 2.4599201224202888e-05, 'epoch': 0.07}) +[2026-06-29 16:20:06,848][starVLA.training.train_starvla][INFO] - Step 166, Loss: {'train/loss': 0.1968042254447937, 'train/grad_norm_pre_clip': 6.440248489379883, 'timing/data': 0.00025035301223397255, 'timing/model': 1.2230678210034966, 'learning_rate/action_model': 9.837742390152232e-05, 'learning_rate/action_query_proj': 9.837742390152232e-05, 'learning_rate/base': 2.459435597538058e-05, 'epoch': 0.07}) +[2026-06-29 16:20:39,192][starVLA.training.train_starvla][INFO] - Step 167, Loss: {'train/loss': 0.6423652172088623, 'train/grad_norm_pre_clip': 3.523211717605591, 'timing/data': 0.00019856402650475502, 'timing/model': 1.221671009901911, 'learning_rate/action_model': 9.835792847454984e-05, 'learning_rate/action_query_proj': 9.835792847454984e-05, 'learning_rate/base': 2.458948211863746e-05, 'epoch': 0.07}) +[2026-06-29 16:21:11,537][starVLA.training.train_starvla][INFO] - Step 168, Loss: {'train/loss': 0.7454636693000793, 'train/grad_norm_pre_clip': 3.217115879058838, 'timing/data': 0.00019925599917769432, 'timing/model': 1.2225472659338266, 'learning_rate/action_model': 9.833831866399716e-05, 'learning_rate/action_query_proj': 9.833831866399716e-05, 'learning_rate/base': 2.458457966599929e-05, 'epoch': 0.07}) +[2026-06-29 16:21:43,889][starVLA.training.train_starvla][INFO] - Step 169, Loss: {'train/loss': 0.21911832690238953, 'train/grad_norm_pre_clip': 5.524785995483398, 'timing/data': 0.0001858309842646122, 'timing/model': 1.2243044250644743, 'learning_rate/action_model': 9.831859451824952e-05, 'learning_rate/action_query_proj': 9.831859451824952e-05, 'learning_rate/base': 2.457964862956238e-05, 'epoch': 0.07}) +[2026-06-29 16:22:16,264][starVLA.training.train_starvla][INFO] - Step 170, Loss: {'train/loss': 0.425038605928421, 'train/grad_norm_pre_clip': 6.034459114074707, 'timing/data': 0.00029015704058110714, 'timing/model': 1.2217038350645453, 'learning_rate/action_model': 9.82987560859743e-05, 'learning_rate/action_query_proj': 9.82987560859743e-05, 'learning_rate/base': 2.4574689021493577e-05, 'epoch': 0.07}) +[2026-06-29 16:22:48,648][starVLA.training.train_starvla][INFO] - Step 171, Loss: {'train/loss': 0.27636808156967163, 'train/grad_norm_pre_clip': 2.234203338623047, 'timing/data': 0.00022618588991463184, 'timing/model': 1.2221327398438007, 'learning_rate/action_model': 9.827880341612087e-05, 'learning_rate/action_query_proj': 9.827880341612087e-05, 'learning_rate/base': 2.4569700854030218e-05, 'epoch': 0.07}) +[2026-06-29 16:23:21,010][starVLA.training.train_starvla][INFO] - Step 172, Loss: {'train/loss': 0.13244251906871796, 'train/grad_norm_pre_clip': 2.431654930114746, 'timing/data': 0.00020234216935932636, 'timing/model': 1.219175579957664, 'learning_rate/action_model': 9.825873655792047e-05, 'learning_rate/action_query_proj': 9.825873655792047e-05, 'learning_rate/base': 2.4564684139480118e-05, 'epoch': 0.07}) +[2026-06-29 16:23:53,506][starVLA.training.train_starvla][INFO] - Step 173, Loss: {'train/loss': 0.06580506265163422, 'train/grad_norm_pre_clip': 2.545114040374756, 'timing/data': 0.00019246293231844902, 'timing/model': 1.2221789530012757, 'learning_rate/action_model': 9.823855556088605e-05, 'learning_rate/action_query_proj': 9.823855556088605e-05, 'learning_rate/base': 2.4559638890221512e-05, 'epoch': 0.07}) +[2026-06-29 16:24:25,838][starVLA.training.train_starvla][INFO] - Step 174, Loss: {'train/loss': 0.00022989464923739433, 'train/grad_norm_pre_clip': 3.72165584564209, 'timing/data': 0.0002747178077697754, 'timing/model': 1.2233046451583505, 'learning_rate/action_model': 9.821826047481222e-05, 'learning_rate/action_query_proj': 9.821826047481222e-05, 'learning_rate/base': 2.4554565118703056e-05, 'epoch': 0.07}) +[2026-06-29 16:24:58,183][starVLA.training.train_starvla][INFO] - Step 175, Loss: {'train/loss': 0.03746525198221207, 'train/grad_norm_pre_clip': 4.182034015655518, 'timing/data': 0.0002490489277988672, 'timing/model': 1.2182146839331836, 'learning_rate/action_model': 9.819785134977507e-05, 'learning_rate/action_query_proj': 9.819785134977507e-05, 'learning_rate/base': 2.4549462837443767e-05, 'epoch': 0.08}) +[2026-06-29 16:25:30,508][starVLA.training.train_starvla][INFO] - Step 176, Loss: {'train/loss': 0.7336449027061462, 'train/grad_norm_pre_clip': 3.1206107139587402, 'timing/data': 0.00019385619089007378, 'timing/model': 1.2210248759947717, 'learning_rate/action_model': 9.817732823613213e-05, 'learning_rate/action_query_proj': 9.817732823613213e-05, 'learning_rate/base': 2.4544332059033032e-05, 'epoch': 0.08}) +[2026-06-29 16:26:02,835][starVLA.training.train_starvla][INFO] - Step 177, Loss: {'train/loss': 0.01867789402604103, 'train/grad_norm_pre_clip': 2.157074213027954, 'timing/data': 0.000352226197719574, 'timing/model': 1.2217456919606775, 'learning_rate/action_model': 9.81566911845221e-05, 'learning_rate/action_query_proj': 9.81566911845221e-05, 'learning_rate/base': 2.4539172796130527e-05, 'epoch': 0.08}) +[2026-06-29 16:26:35,161][starVLA.training.train_starvla][INFO] - Step 178, Loss: {'train/loss': 0.012784660793840885, 'train/grad_norm_pre_clip': 3.1366753578186035, 'timing/data': 0.00019147200509905815, 'timing/model': 1.2205308810807765, 'learning_rate/action_model': 9.81359402458649e-05, 'learning_rate/action_query_proj': 9.81359402458649e-05, 'learning_rate/base': 2.4533985061466225e-05, 'epoch': 0.08}) +[2026-06-29 16:27:07,503][starVLA.training.train_starvla][INFO] - Step 179, Loss: {'train/loss': 0.1803593635559082, 'train/grad_norm_pre_clip': 1.7226780652999878, 'timing/data': 0.00025025196373462677, 'timing/model': 1.2194322298746556, 'learning_rate/action_model': 9.811507547136136e-05, 'learning_rate/action_query_proj': 9.811507547136136e-05, 'learning_rate/base': 2.452876886784034e-05, 'epoch': 0.08}) +[2026-06-29 16:27:39,794][starVLA.training.train_starvla][INFO] - Step 180, Loss: {'train/loss': 0.09992855787277222, 'train/grad_norm_pre_clip': 2.3835129737854004, 'timing/data': 0.00019316491670906544, 'timing/model': 1.221850223839283, 'learning_rate/action_model': 9.809409691249328e-05, 'learning_rate/action_query_proj': 9.809409691249328e-05, 'learning_rate/base': 2.452352422812332e-05, 'epoch': 0.08}) +[2026-06-29 16:28:12,201][starVLA.training.train_starvla][INFO] - Step 181, Loss: {'train/loss': 0.013033845461905003, 'train/grad_norm_pre_clip': 3.1720070838928223, 'timing/data': 0.00019188201986253262, 'timing/model': 1.2240997380577028, 'learning_rate/action_model': 9.807300462102312e-05, 'learning_rate/action_query_proj': 9.807300462102312e-05, 'learning_rate/base': 2.451825115525578e-05, 'epoch': 0.08}) +[2026-06-29 16:28:44,540][starVLA.training.train_starvla][INFO] - Step 182, Loss: {'train/loss': 0.07623562216758728, 'train/grad_norm_pre_clip': 2.745760440826416, 'timing/data': 0.00025093392468988895, 'timing/model': 1.211480907863006, 'learning_rate/action_model': 9.805179864899407e-05, 'learning_rate/action_query_proj': 9.805179864899407e-05, 'learning_rate/base': 2.4512949662248516e-05, 'epoch': 0.08}) +[2026-06-29 16:29:16,839][starVLA.training.train_starvla][INFO] - Step 183, Loss: {'train/loss': 0.23545540869235992, 'train/grad_norm_pre_clip': 4.019288539886475, 'timing/data': 0.00023967213928699493, 'timing/model': 1.2230063858442008, 'learning_rate/action_model': 9.80304790487297e-05, 'learning_rate/action_query_proj': 9.80304790487297e-05, 'learning_rate/base': 2.4507619762182426e-05, 'epoch': 0.08}) +[2026-06-29 16:29:49,204][starVLA.training.train_starvla][INFO] - Step 184, Loss: {'train/loss': 0.033456798642873764, 'train/grad_norm_pre_clip': 3.557187080383301, 'timing/data': 0.00019568903371691704, 'timing/model': 1.2197969919070601, 'learning_rate/action_model': 9.800904587283404e-05, 'learning_rate/action_query_proj': 9.800904587283404e-05, 'learning_rate/base': 2.450226146820851e-05, 'epoch': 0.08}) +[2026-06-29 16:30:21,549][starVLA.training.train_starvla][INFO] - Step 185, Loss: {'train/loss': 0.008165678940713406, 'train/grad_norm_pre_clip': 1.8721210956573486, 'timing/data': 0.0002464749850332737, 'timing/model': 1.2194010878447443, 'learning_rate/action_model': 9.798749917419133e-05, 'learning_rate/action_query_proj': 9.798749917419133e-05, 'learning_rate/base': 2.4496874793547833e-05, 'epoch': 0.08}) +[2026-06-29 16:30:53,881][starVLA.training.train_starvla][INFO] - Step 186, Loss: {'train/loss': 0.01291560661047697, 'train/grad_norm_pre_clip': 3.3906590938568115, 'timing/data': 0.00023823999799787998, 'timing/model': 1.2216642608400434, 'learning_rate/action_model': 9.796583900596586e-05, 'learning_rate/action_query_proj': 9.796583900596586e-05, 'learning_rate/base': 2.4491459751491465e-05, 'epoch': 0.08}) +[2026-06-29 16:31:26,189][starVLA.training.train_starvla][INFO] - Step 187, Loss: {'train/loss': 0.14614081382751465, 'train/grad_norm_pre_clip': 5.031567096710205, 'timing/data': 0.00024118483997881413, 'timing/model': 1.2234842940233648, 'learning_rate/action_model': 9.794406542160201e-05, 'learning_rate/action_query_proj': 9.794406542160201e-05, 'learning_rate/base': 2.4486016355400503e-05, 'epoch': 0.08}) +[2026-06-29 16:31:58,559][starVLA.training.train_starvla][INFO] - Step 188, Loss: {'train/loss': 0.27414944767951965, 'train/grad_norm_pre_clip': 1.813660979270935, 'timing/data': 0.0002598999999463558, 'timing/model': 1.2178642419166863, 'learning_rate/action_model': 9.792217847482387e-05, 'learning_rate/action_query_proj': 9.792217847482387e-05, 'learning_rate/base': 2.4480544618705968e-05, 'epoch': 0.08}) +[2026-06-29 16:32:30,920][starVLA.training.train_starvla][INFO] - Step 189, Loss: {'train/loss': 0.12840500473976135, 'train/grad_norm_pre_clip': 5.154791831970215, 'timing/data': 0.0002877330407500267, 'timing/model': 1.2181744491681457, 'learning_rate/action_model': 9.790017821963533e-05, 'learning_rate/action_query_proj': 9.790017821963533e-05, 'learning_rate/base': 2.4475044554908833e-05, 'epoch': 0.08}) +[2026-06-29 16:33:03,299][starVLA.training.train_starvla][INFO] - Step 190, Loss: {'train/loss': 0.017338894307613373, 'train/grad_norm_pre_clip': 2.682182550430298, 'timing/data': 0.00026194402016699314, 'timing/model': 1.2213944639079273, 'learning_rate/action_model': 9.787806471031985e-05, 'learning_rate/action_query_proj': 9.787806471031985e-05, 'learning_rate/base': 2.4469516177579963e-05, 'epoch': 0.08}) +[2026-06-29 16:33:35,641][starVLA.training.train_starvla][INFO] - Step 191, Loss: {'train/loss': 0.2760661840438843, 'train/grad_norm_pre_clip': 1.1897464990615845, 'timing/data': 0.00020873406901955605, 'timing/model': 1.2217941721901298, 'learning_rate/action_model': 9.78558380014403e-05, 'learning_rate/action_query_proj': 9.78558380014403e-05, 'learning_rate/base': 2.4463959500360074e-05, 'epoch': 0.08}) +[2026-06-29 16:34:07,956][starVLA.training.train_starvla][INFO] - Step 192, Loss: {'train/loss': 0.526630163192749, 'train/grad_norm_pre_clip': 1.9448596239089966, 'timing/data': 0.0001999379601329565, 'timing/model': 1.2118128428701311, 'learning_rate/action_model': 9.783349814783888e-05, 'learning_rate/action_query_proj': 9.783349814783888e-05, 'learning_rate/base': 2.445837453695972e-05, 'epoch': 0.08}) +[2026-06-29 16:34:40,278][starVLA.training.train_starvla][INFO] - Step 193, Loss: {'train/loss': 0.005100904032588005, 'train/grad_norm_pre_clip': 2.607710599899292, 'timing/data': 0.00019890489056706429, 'timing/model': 1.220995815936476, 'learning_rate/action_model': 9.781104520463692e-05, 'learning_rate/action_query_proj': 9.781104520463692e-05, 'learning_rate/base': 2.445276130115923e-05, 'epoch': 0.08}) +[2026-06-29 16:35:12,617][starVLA.training.train_starvla][INFO] - Step 194, Loss: {'train/loss': 0.09166241437196732, 'train/grad_norm_pre_clip': 2.193223237991333, 'timing/data': 0.000291160074993968, 'timing/model': 1.2171471319161355, 'learning_rate/action_model': 9.77884792272349e-05, 'learning_rate/action_query_proj': 9.77884792272349e-05, 'learning_rate/base': 2.4447119806808725e-05, 'epoch': 0.08}) +[2026-06-29 16:35:44,997][starVLA.training.train_starvla][INFO] - Step 195, Loss: {'train/loss': 0.07388582080602646, 'train/grad_norm_pre_clip': 8.71343994140625, 'timing/data': 0.00019435607828199863, 'timing/model': 1.2210638360120356, 'learning_rate/action_model': 9.776580027131206e-05, 'learning_rate/action_query_proj': 9.776580027131206e-05, 'learning_rate/base': 2.4441450067828016e-05, 'epoch': 0.08}) +[2026-06-29 16:36:17,337][starVLA.training.train_starvla][INFO] - Step 196, Loss: {'train/loss': 0.014347673393785954, 'train/grad_norm_pre_clip': 3.922208309173584, 'timing/data': 0.0002906189765781164, 'timing/model': 1.2171167449560016, 'learning_rate/action_model': 9.774300839282652e-05, 'learning_rate/action_query_proj': 9.774300839282652e-05, 'learning_rate/base': 2.443575209820663e-05, 'epoch': 0.08}) +[2026-06-29 16:36:49,665][starVLA.training.train_starvla][INFO] - Step 197, Loss: {'train/loss': 0.6890566349029541, 'train/grad_norm_pre_clip': 3.0419092178344727, 'timing/data': 0.0003058768343180418, 'timing/model': 1.2229898280929774, 'learning_rate/action_model': 9.772010364801491e-05, 'learning_rate/action_query_proj': 9.772010364801491e-05, 'learning_rate/base': 2.443002591200373e-05, 'epoch': 0.08}) +[2026-06-29 16:37:22,023][starVLA.training.train_starvla][INFO] - Step 198, Loss: {'train/loss': 0.5154087543487549, 'train/grad_norm_pre_clip': 2.0237841606140137, 'timing/data': 0.0001932650338858366, 'timing/model': 1.2212631450966, 'learning_rate/action_model': 9.769708609339249e-05, 'learning_rate/action_query_proj': 9.769708609339249e-05, 'learning_rate/base': 2.4424271523348122e-05, 'epoch': 0.09}) +[2026-06-29 16:37:54,410][starVLA.training.train_starvla][INFO] - Step 199, Loss: {'train/loss': 0.004981061909347773, 'train/grad_norm_pre_clip': 2.4004640579223633, 'timing/data': 0.00032146694138646126, 'timing/model': 1.2205616170540452, 'learning_rate/action_model': 9.767395578575272e-05, 'learning_rate/action_query_proj': 9.767395578575272e-05, 'learning_rate/base': 2.441848894643818e-05, 'epoch': 0.09}) +[2026-06-29 16:38:26,747][starVLA.training.train_starvla][INFO] - Step 200, Loss: {'train/loss': 0.04218661040067673, 'train/grad_norm_pre_clip': 3.3669586181640625, 'timing/data': 0.00029632984660565853, 'timing/model': 1.2183799322228879, 'learning_rate/action_model': 9.765071278216737e-05, 'learning_rate/action_query_proj': 9.765071278216737e-05, 'learning_rate/base': 2.4412678195541843e-05, 'epoch': 0.09}) +[2026-06-29 16:38:59,101][starVLA.training.train_starvla][INFO] - Step 201, Loss: {'train/loss': 0.280961811542511, 'train/grad_norm_pre_clip': 3.673773765563965, 'timing/data': 0.00019670207984745502, 'timing/model': 1.2182136431802064, 'learning_rate/action_model': 9.762735713998625e-05, 'learning_rate/action_query_proj': 9.762735713998625e-05, 'learning_rate/base': 2.4406839284996563e-05, 'epoch': 0.09}) +[2026-06-29 16:39:31,434][starVLA.training.train_starvla][INFO] - Step 202, Loss: {'train/loss': 0.0292922705411911, 'train/grad_norm_pre_clip': 3.6886749267578125, 'timing/data': 0.0002743781078606844, 'timing/model': 1.2185880558099598, 'learning_rate/action_model': 9.760388891683704e-05, 'learning_rate/action_query_proj': 9.760388891683704e-05, 'learning_rate/base': 2.440097222920926e-05, 'epoch': 0.09}) +[2026-06-29 16:40:03,763][starVLA.training.train_starvla][INFO] - Step 203, Loss: {'train/loss': 0.1654965877532959, 'train/grad_norm_pre_clip': 2.493999481201172, 'timing/data': 0.00020724092610180378, 'timing/model': 1.2189728408120573, 'learning_rate/action_model': 9.75803081706253e-05, 'learning_rate/action_query_proj': 9.75803081706253e-05, 'learning_rate/base': 2.4395077042656325e-05, 'epoch': 0.09}) +[2026-06-29 16:40:36,120][starVLA.training.train_starvla][INFO] - Step 204, Loss: {'train/loss': 0.7247751951217651, 'train/grad_norm_pre_clip': 4.044231414794922, 'timing/data': 0.00018980796448886395, 'timing/model': 1.22010234487243, 'learning_rate/action_model': 9.755661495953414e-05, 'learning_rate/action_query_proj': 9.755661495953414e-05, 'learning_rate/base': 2.4389153739883535e-05, 'epoch': 0.09}) +[2026-06-29 16:41:08,498][starVLA.training.train_starvla][INFO] - Step 205, Loss: {'train/loss': 0.07767203450202942, 'train/grad_norm_pre_clip': 6.139837265014648, 'timing/data': 0.00024013291113078594, 'timing/model': 1.2219521990045905, 'learning_rate/action_model': 9.753280934202423e-05, 'learning_rate/action_query_proj': 9.753280934202423e-05, 'learning_rate/base': 2.4383202335506057e-05, 'epoch': 0.09}) +[2026-06-29 16:41:40,864][starVLA.training.train_starvla][INFO] - Step 206, Loss: {'train/loss': 0.0019981253426522017, 'train/grad_norm_pre_clip': 4.6727728843688965, 'timing/data': 0.0002498610410839319, 'timing/model': 1.2189763511996716, 'learning_rate/action_model': 9.750889137683354e-05, 'learning_rate/action_query_proj': 9.750889137683354e-05, 'learning_rate/base': 2.4377222844208385e-05, 'epoch': 0.09}) +[2026-06-29 16:42:13,222][starVLA.training.train_starvla][INFO] - Step 207, Loss: {'train/loss': 0.08848702907562256, 'train/grad_norm_pre_clip': 3.3031020164489746, 'timing/data': 0.00025687413290143013, 'timing/model': 1.2239994260016829, 'learning_rate/action_model': 9.748486112297729e-05, 'learning_rate/action_query_proj': 9.748486112297729e-05, 'learning_rate/base': 2.4371215280744323e-05, 'epoch': 0.09}) +[2026-06-29 16:42:45,552][starVLA.training.train_starvla][INFO] - Step 208, Loss: {'train/loss': 0.4405851662158966, 'train/grad_norm_pre_clip': 3.6030962467193604, 'timing/data': 0.0001992359757423401, 'timing/model': 1.2194470579270273, 'learning_rate/action_model': 9.746071863974773e-05, 'learning_rate/action_query_proj': 9.746071863974773e-05, 'learning_rate/base': 2.4365179659936933e-05, 'epoch': 0.09}) +[2026-06-29 16:43:17,879][starVLA.training.train_starvla][INFO] - Step 209, Loss: {'train/loss': 0.4188750684261322, 'train/grad_norm_pre_clip': 2.4011926651000977, 'timing/data': 0.00025504082441329956, 'timing/model': 1.2194884188938886, 'learning_rate/action_model': 9.743646398671405e-05, 'learning_rate/action_query_proj': 9.743646398671405e-05, 'learning_rate/base': 2.4359115996678512e-05, 'epoch': 0.09}) +[2026-06-29 16:43:50,229][starVLA.training.train_starvla][INFO] - Step 210, Loss: {'train/loss': 0.16502895951271057, 'train/grad_norm_pre_clip': 4.836208343505859, 'timing/data': 0.00019128108397126198, 'timing/model': 1.2245395919308066, 'learning_rate/action_model': 9.741209722372218e-05, 'learning_rate/action_query_proj': 9.741209722372218e-05, 'learning_rate/base': 2.4353024305930544e-05, 'epoch': 0.09}) +[2026-06-29 16:44:22,581][starVLA.training.train_starvla][INFO] - Step 211, Loss: {'train/loss': 0.00559856416657567, 'train/grad_norm_pre_clip': 8.430062294006348, 'timing/data': 0.00019370601512491703, 'timing/model': 1.2206812479998916, 'learning_rate/action_model': 9.73876184108947e-05, 'learning_rate/action_query_proj': 9.73876184108947e-05, 'learning_rate/base': 2.4346904602723675e-05, 'epoch': 0.09}) +[2026-06-29 16:44:55,063][starVLA.training.train_starvla][INFO] - Step 212, Loss: {'train/loss': 0.06922651827335358, 'train/grad_norm_pre_clip': 5.250257968902588, 'timing/data': 0.0003092139959335327, 'timing/model': 1.2190311618614942, 'learning_rate/action_model': 9.736302760863063e-05, 'learning_rate/action_query_proj': 9.736302760863063e-05, 'learning_rate/base': 2.4340756902157657e-05, 'epoch': 0.09}) +[2026-06-29 16:45:27,396][starVLA.training.train_starvla][INFO] - Step 213, Loss: {'train/loss': 0.22668519616127014, 'train/grad_norm_pre_clip': 4.030121326446533, 'timing/data': 0.00019015907309949398, 'timing/model': 1.2194947651587427, 'learning_rate/action_model': 9.733832487760536e-05, 'learning_rate/action_query_proj': 9.733832487760536e-05, 'learning_rate/base': 2.433458121940134e-05, 'epoch': 0.09}) +[2026-06-29 16:45:59,771][starVLA.training.train_starvla][INFO] - Step 214, Loss: {'train/loss': 0.17097492516040802, 'train/grad_norm_pre_clip': 4.077707290649414, 'timing/data': 0.0002091941423714161, 'timing/model': 1.2116257199086249, 'learning_rate/action_model': 9.73135102787704e-05, 'learning_rate/action_query_proj': 9.73135102787704e-05, 'learning_rate/base': 2.43283775696926e-05, 'epoch': 0.09}) +[2026-06-29 16:46:32,119][starVLA.training.train_starvla][INFO] - Step 215, Loss: {'train/loss': 0.0689110979437828, 'train/grad_norm_pre_clip': 3.8733015060424805, 'timing/data': 0.00019437703303992748, 'timing/model': 1.2228521502111107, 'learning_rate/action_model': 9.72885838733533e-05, 'learning_rate/action_query_proj': 9.72885838733533e-05, 'learning_rate/base': 2.4322145968338326e-05, 'epoch': 0.09}) +[2026-06-29 16:47:04,494][starVLA.training.train_starvla][INFO] - Step 216, Loss: {'train/loss': 0.27316609025001526, 'train/grad_norm_pre_clip': 1.2936842441558838, 'timing/data': 0.00020031793974339962, 'timing/model': 1.222644800087437, 'learning_rate/action_model': 9.72635457228575e-05, 'learning_rate/action_query_proj': 9.72635457228575e-05, 'learning_rate/base': 2.4315886430714374e-05, 'epoch': 0.09}) +[2026-06-29 16:47:36,861][starVLA.training.train_starvla][INFO] - Step 217, Loss: {'train/loss': 0.013750198297202587, 'train/grad_norm_pre_clip': 1.731759786605835, 'timing/data': 0.00019504805095493793, 'timing/model': 1.2187185469083488, 'learning_rate/action_model': 9.723839588906215e-05, 'learning_rate/action_query_proj': 9.723839588906215e-05, 'learning_rate/base': 2.4309598972265538e-05, 'epoch': 0.09}) +[2026-06-29 16:48:09,215][starVLA.training.train_starvla][INFO] - Step 218, Loss: {'train/loss': 0.17553627490997314, 'train/grad_norm_pre_clip': 3.3576626777648926, 'timing/data': 0.00023837899789214134, 'timing/model': 1.2195879311766475, 'learning_rate/action_model': 9.721313443402194e-05, 'learning_rate/action_query_proj': 9.721313443402194e-05, 'learning_rate/base': 2.4303283608505486e-05, 'epoch': 0.09}) +[2026-06-29 16:48:41,573][starVLA.training.train_starvla][INFO] - Step 219, Loss: {'train/loss': 0.0611262246966362, 'train/grad_norm_pre_clip': 3.892277717590332, 'timing/data': 0.00018900702707469463, 'timing/model': 1.2255671159364283, 'learning_rate/action_model': 9.718776142006706e-05, 'learning_rate/action_query_proj': 9.718776142006706e-05, 'learning_rate/base': 2.4296940355016764e-05, 'epoch': 0.09}) +[2026-06-29 16:49:13,934][starVLA.training.train_starvla][INFO] - Step 220, Loss: {'train/loss': 1.9127165079116821, 'train/grad_norm_pre_clip': 2.7637712955474854, 'timing/data': 0.0003074798732995987, 'timing/model': 1.2235609968192875, 'learning_rate/action_model': 9.716227690980283e-05, 'learning_rate/action_query_proj': 9.716227690980283e-05, 'learning_rate/base': 2.4290569227450707e-05, 'epoch': 0.09}) +[2026-06-29 16:49:46,266][starVLA.training.train_starvla][INFO] - Step 221, Loss: {'train/loss': 0.048252612352371216, 'train/grad_norm_pre_clip': 2.9768784046173096, 'timing/data': 0.00020165112800896168, 'timing/model': 1.219530340982601, 'learning_rate/action_model': 9.713668096610978e-05, 'learning_rate/action_query_proj': 9.713668096610978e-05, 'learning_rate/base': 2.4284170241527445e-05, 'epoch': 0.1}) +[2026-06-29 16:50:18,624][starVLA.training.train_starvla][INFO] - Step 222, Loss: {'train/loss': 0.3762499988079071, 'train/grad_norm_pre_clip': 1.6035141944885254, 'timing/data': 0.00018676207400858402, 'timing/model': 1.222125044092536, 'learning_rate/action_model': 9.711097365214335e-05, 'learning_rate/action_query_proj': 9.711097365214335e-05, 'learning_rate/base': 2.4277743413035837e-05, 'epoch': 0.1}) +[2026-06-29 16:50:51,010][starVLA.training.train_starvla][INFO] - Step 223, Loss: {'train/loss': 0.6012417674064636, 'train/grad_norm_pre_clip': 2.9006505012512207, 'timing/data': 0.00019682105630636215, 'timing/model': 1.2191874210257083, 'learning_rate/action_model': 9.708515503133378e-05, 'learning_rate/action_query_proj': 9.708515503133378e-05, 'learning_rate/base': 2.4271288757833445e-05, 'epoch': 0.1}) +[2026-06-29 16:51:23,372][starVLA.training.train_starvla][INFO] - Step 224, Loss: {'train/loss': 0.004535223823040724, 'train/grad_norm_pre_clip': 3.555832862854004, 'timing/data': 0.0003033420071005821, 'timing/model': 1.22378841903992, 'learning_rate/action_model': 9.705922516738596e-05, 'learning_rate/action_query_proj': 9.705922516738596e-05, 'learning_rate/base': 2.426480629184649e-05, 'epoch': 0.1}) +[2026-06-29 16:51:55,717][starVLA.training.train_starvla][INFO] - Step 225, Loss: {'train/loss': 0.0018608818063512444, 'train/grad_norm_pre_clip': 5.209141731262207, 'timing/data': 0.00020219082944095135, 'timing/model': 1.212100180098787, 'learning_rate/action_model': 9.703318412427925e-05, 'learning_rate/action_query_proj': 9.703318412427925e-05, 'learning_rate/base': 2.4258296031069812e-05, 'epoch': 0.1}) +[2026-06-29 16:52:28,030][starVLA.training.train_starvla][INFO] - Step 226, Loss: {'train/loss': 0.13068929314613342, 'train/grad_norm_pre_clip': 3.1590170860290527, 'timing/data': 0.0001900091301649809, 'timing/model': 1.225027927896008, 'learning_rate/action_model': 9.700703196626731e-05, 'learning_rate/action_query_proj': 9.700703196626731e-05, 'learning_rate/base': 2.4251757991566827e-05, 'epoch': 0.1}) +[2026-06-29 16:53:00,306][starVLA.training.train_starvla][INFO] - Step 227, Loss: {'train/loss': 0.025729238986968994, 'train/grad_norm_pre_clip': 2.3559141159057617, 'timing/data': 0.00020476593635976315, 'timing/model': 1.2231075060553849, 'learning_rate/action_model': 9.698076875787803e-05, 'learning_rate/action_query_proj': 9.698076875787803e-05, 'learning_rate/base': 2.4245192189469508e-05, 'epoch': 0.1}) +[2026-06-29 16:53:32,666][starVLA.training.train_starvla][INFO] - Step 228, Loss: {'train/loss': 7.837978955649305e-06, 'train/grad_norm_pre_clip': 3.7165420055389404, 'timing/data': 0.00019542896188795567, 'timing/model': 1.2220107689499855, 'learning_rate/action_model': 9.695439456391323e-05, 'learning_rate/action_query_proj': 9.695439456391323e-05, 'learning_rate/base': 2.4238598640978308e-05, 'epoch': 0.1}) +[2026-06-29 16:54:05,036][starVLA.training.train_starvla][INFO] - Step 229, Loss: {'train/loss': 0.09998033940792084, 'train/grad_norm_pre_clip': 2.1265382766723633, 'timing/data': 0.00026854686439037323, 'timing/model': 1.2183021649252623, 'learning_rate/action_model': 9.692790944944865e-05, 'learning_rate/action_query_proj': 9.692790944944865e-05, 'learning_rate/base': 2.4231977362362163e-05, 'epoch': 0.1}) +[2026-06-29 16:54:37,400][starVLA.training.train_starvla][INFO] - Step 230, Loss: {'train/loss': 0.12204325199127197, 'train/grad_norm_pre_clip': 2.1644530296325684, 'timing/data': 0.0002531569916754961, 'timing/model': 1.2211528250481933, 'learning_rate/action_model': 9.690131347983364e-05, 'learning_rate/action_query_proj': 9.690131347983364e-05, 'learning_rate/base': 2.422532836995841e-05, 'epoch': 0.1}) +[2026-06-29 16:55:09,722][starVLA.training.train_starvla][INFO] - Step 231, Loss: {'train/loss': 0.001162309548817575, 'train/grad_norm_pre_clip': 2.186314105987549, 'timing/data': 0.00020158011466264725, 'timing/model': 1.217958854045719, 'learning_rate/action_model': 9.687460672069112e-05, 'learning_rate/action_query_proj': 9.687460672069112e-05, 'learning_rate/base': 2.421865168017278e-05, 'epoch': 0.1}) +[2026-06-29 16:55:42,033][starVLA.training.train_starvla][INFO] - Step 232, Loss: {'train/loss': 0.006960941478610039, 'train/grad_norm_pre_clip': 4.385502815246582, 'timing/data': 0.00027711200527846813, 'timing/model': 1.2204305690247566, 'learning_rate/action_model': 9.684778923791739e-05, 'learning_rate/action_query_proj': 9.684778923791739e-05, 'learning_rate/base': 2.4211947309479347e-05, 'epoch': 0.1}) +[2026-06-29 16:56:14,411][starVLA.training.train_starvla][INFO] - Step 233, Loss: {'train/loss': 0.0830448567867279, 'train/grad_norm_pre_clip': 6.408285617828369, 'timing/data': 0.000298523111268878, 'timing/model': 1.2324003630783409, 'learning_rate/action_model': 9.682086109768189e-05, 'learning_rate/action_query_proj': 9.682086109768189e-05, 'learning_rate/base': 2.420521527442047e-05, 'epoch': 0.1}) +[2026-06-29 16:56:46,768][starVLA.training.train_starvla][INFO] - Step 234, Loss: {'train/loss': 0.04836129769682884, 'train/grad_norm_pre_clip': 4.080545425415039, 'timing/data': 0.00020110909827053547, 'timing/model': 1.221064873971045, 'learning_rate/action_model': 9.679382236642713e-05, 'learning_rate/action_query_proj': 9.679382236642713e-05, 'learning_rate/base': 2.4198455591606783e-05, 'epoch': 0.1}) +[2026-06-29 16:57:19,146][starVLA.training.train_starvla][INFO] - Step 235, Loss: {'train/loss': 0.017321698367595673, 'train/grad_norm_pre_clip': 2.0600857734680176, 'timing/data': 0.00019673118367791176, 'timing/model': 1.2200358370319009, 'learning_rate/action_model': 9.676667311086852e-05, 'learning_rate/action_query_proj': 9.676667311086852e-05, 'learning_rate/base': 2.419166827771713e-05, 'epoch': 0.1}) +[2026-06-29 16:57:51,513][starVLA.training.train_starvla][INFO] - Step 236, Loss: {'train/loss': 0.37067222595214844, 'train/grad_norm_pre_clip': 3.1458423137664795, 'timing/data': 0.00023358012549579144, 'timing/model': 1.2205858200322837, 'learning_rate/action_model': 9.673941339799412e-05, 'learning_rate/action_query_proj': 9.673941339799412e-05, 'learning_rate/base': 2.418485334949853e-05, 'epoch': 0.1}) +[2026-06-29 16:58:23,867][starVLA.training.train_starvla][INFO] - Step 237, Loss: {'train/loss': 0.5219566226005554, 'train/grad_norm_pre_clip': 1.7510749101638794, 'timing/data': 0.000248519005253911, 'timing/model': 1.2216650710906833, 'learning_rate/action_model': 9.671204329506459e-05, 'learning_rate/action_query_proj': 9.671204329506459e-05, 'learning_rate/base': 2.4178010823766147e-05, 'epoch': 0.1}) +[2026-06-29 16:58:56,228][starVLA.training.train_starvla][INFO] - Step 238, Loss: {'train/loss': 0.021422000601887703, 'train/grad_norm_pre_clip': 2.736851215362549, 'timing/data': 0.0001889960840344429, 'timing/model': 1.2175840679556131, 'learning_rate/action_model': 9.668456286961291e-05, 'learning_rate/action_query_proj': 9.668456286961291e-05, 'learning_rate/base': 2.4171140717403227e-05, 'epoch': 0.1}) +[2026-06-29 16:59:28,613][starVLA.training.train_starvla][INFO] - Step 239, Loss: {'train/loss': 0.11514803022146225, 'train/grad_norm_pre_clip': 2.8335623741149902, 'timing/data': 0.00036892620846629143, 'timing/model': 1.2201725279446691, 'learning_rate/action_model': 9.665697218944432e-05, 'learning_rate/action_query_proj': 9.665697218944432e-05, 'learning_rate/base': 2.416424304736108e-05, 'epoch': 0.1}) +[2026-06-29 17:00:01,017][starVLA.training.train_starvla][INFO] - Step 240, Loss: {'train/loss': 0.018496057018637657, 'train/grad_norm_pre_clip': 2.863872766494751, 'timing/data': 0.00028787506744265556, 'timing/model': 1.2217942539136857, 'learning_rate/action_model': 9.662927132263607e-05, 'learning_rate/action_query_proj': 9.662927132263607e-05, 'learning_rate/base': 2.4157317830659017e-05, 'epoch': 0.1}) +[2026-06-29 17:00:33,356][starVLA.training.train_starvla][INFO] - Step 241, Loss: {'train/loss': 0.1940646767616272, 'train/grad_norm_pre_clip': 3.1331615447998047, 'timing/data': 0.0002396728377789259, 'timing/model': 1.218735315836966, 'learning_rate/action_model': 9.66014603375373e-05, 'learning_rate/action_query_proj': 9.66014603375373e-05, 'learning_rate/base': 2.4150365084384325e-05, 'epoch': 0.1}) +[2026-06-29 17:01:05,703][starVLA.training.train_starvla][INFO] - Step 242, Loss: {'train/loss': 0.0040220823138952255, 'train/grad_norm_pre_clip': 1.6070698499679565, 'timing/data': 0.00020248210057616234, 'timing/model': 1.2207373108249158, 'learning_rate/action_model': 9.657353930276884e-05, 'learning_rate/action_query_proj': 9.657353930276884e-05, 'learning_rate/base': 2.414338482569221e-05, 'epoch': 0.1}) +[2026-06-29 17:01:38,046][starVLA.training.train_starvla][INFO] - Step 243, Loss: {'train/loss': 0.019510919228196144, 'train/grad_norm_pre_clip': 3.0859177112579346, 'timing/data': 0.0002471769694238901, 'timing/model': 1.22158925794065, 'learning_rate/action_model': 9.654550828722308e-05, 'learning_rate/action_query_proj': 9.654550828722308e-05, 'learning_rate/base': 2.413637707180577e-05, 'epoch': 0.1}) +[2026-06-29 17:02:10,396][starVLA.training.train_starvla][INFO] - Step 244, Loss: {'train/loss': 0.07132203876972198, 'train/grad_norm_pre_clip': 2.661186456680298, 'timing/data': 0.00023192795924842358, 'timing/model': 1.2199634548742324, 'learning_rate/action_model': 9.651736736006375e-05, 'learning_rate/action_query_proj': 9.651736736006375e-05, 'learning_rate/base': 2.4129341840015938e-05, 'epoch': 0.1}) +[2026-06-29 17:02:42,729][starVLA.training.train_starvla][INFO] - Step 245, Loss: {'train/loss': 0.006472217384725809, 'train/grad_norm_pre_clip': 2.671596050262451, 'timing/data': 0.00027421792037785053, 'timing/model': 1.2201281099114567, 'learning_rate/action_model': 9.648911659072579e-05, 'learning_rate/action_query_proj': 9.648911659072579e-05, 'learning_rate/base': 2.4122279147681448e-05, 'epoch': 0.11}) +[2026-06-29 17:03:15,053][starVLA.training.train_starvla][INFO] - Step 246, Loss: {'train/loss': 0.037211786955595016, 'train/grad_norm_pre_clip': 2.8874781131744385, 'timing/data': 0.00034494209103286266, 'timing/model': 1.2178481000009924, 'learning_rate/action_model': 9.64607560489152e-05, 'learning_rate/action_query_proj': 9.64607560489152e-05, 'learning_rate/base': 2.41151890122288e-05, 'epoch': 0.11}) +[2026-06-29 17:03:47,397][starVLA.training.train_starvla][INFO] - Step 247, Loss: {'train/loss': 0.05689190700650215, 'train/grad_norm_pre_clip': 1.6316335201263428, 'timing/data': 0.00019632000476121902, 'timing/model': 1.2212863638997078, 'learning_rate/action_model': 9.643228580460877e-05, 'learning_rate/action_query_proj': 9.643228580460877e-05, 'learning_rate/base': 2.4108071451152192e-05, 'epoch': 0.11}) +[2026-06-29 17:04:19,749][starVLA.training.train_starvla][INFO] - Step 248, Loss: {'train/loss': 0.13598224520683289, 'train/grad_norm_pre_clip': 3.6646735668182373, 'timing/data': 0.00024788803420960903, 'timing/model': 1.2189466741401702, 'learning_rate/action_model': 9.640370592805399e-05, 'learning_rate/action_query_proj': 9.640370592805399e-05, 'learning_rate/base': 2.4100926482013497e-05, 'epoch': 0.11}) +[2026-06-29 17:04:52,038][starVLA.training.train_starvla][INFO] - Step 249, Loss: {'train/loss': 0.05876057967543602, 'train/grad_norm_pre_clip': 2.4839720726013184, 'timing/data': 0.00031203008256852627, 'timing/model': 1.220845822012052, 'learning_rate/action_model': 9.637501648976889e-05, 'learning_rate/action_query_proj': 9.637501648976889e-05, 'learning_rate/base': 2.409375412244222e-05, 'epoch': 0.11}) +[2026-06-29 17:05:24,540][starVLA.training.train_starvla][INFO] - Step 250, Loss: {'train/loss': 0.21438190340995789, 'train/grad_norm_pre_clip': 2.2382421493530273, 'timing/data': 0.00020713103003799915, 'timing/model': 1.220438471995294, 'learning_rate/action_model': 9.634621756054177e-05, 'learning_rate/action_query_proj': 9.634621756054177e-05, 'learning_rate/base': 2.4086554390135442e-05, 'epoch': 0.11}) +[2026-06-29 17:05:56,858][starVLA.training.train_starvla][INFO] - Step 251, Loss: {'train/loss': 0.23028656840324402, 'train/grad_norm_pre_clip': 4.792329788208008, 'timing/data': 0.00021562702022492886, 'timing/model': 1.2172946529462934, 'learning_rate/action_model': 9.631730921143115e-05, 'learning_rate/action_query_proj': 9.631730921143115e-05, 'learning_rate/base': 2.4079327302857788e-05, 'epoch': 0.11}) +[2026-06-29 17:06:29,190][starVLA.training.train_starvla][INFO] - Step 252, Loss: {'train/loss': 0.6132359504699707, 'train/grad_norm_pre_clip': 3.606961250305176, 'timing/data': 0.0001951789017766714, 'timing/model': 1.2210830149706453, 'learning_rate/action_model': 9.628829151376552e-05, 'learning_rate/action_query_proj': 9.628829151376552e-05, 'learning_rate/base': 2.407207287844138e-05, 'epoch': 0.11}) +[2026-06-29 17:07:01,582][starVLA.training.train_starvla][INFO] - Step 253, Loss: {'train/loss': 0.10149561613798141, 'train/grad_norm_pre_clip': 1.3736742734909058, 'timing/data': 0.00034804712049663067, 'timing/model': 1.222732936963439, 'learning_rate/action_model': 9.625916453914314e-05, 'learning_rate/action_query_proj': 9.625916453914314e-05, 'learning_rate/base': 2.4064791134785784e-05, 'epoch': 0.11}) +[2026-06-29 17:07:33,947][starVLA.training.train_starvla][INFO] - Step 254, Loss: {'train/loss': 0.013839676044881344, 'train/grad_norm_pre_clip': 1.636723518371582, 'timing/data': 0.00025843805633485317, 'timing/model': 1.2276815450750291, 'learning_rate/action_model': 9.622992835943193e-05, 'learning_rate/action_query_proj': 9.622992835943193e-05, 'learning_rate/base': 2.4057482089857984e-05, 'epoch': 0.11}) +[2026-06-29 17:08:06,285][starVLA.training.train_starvla][INFO] - Step 255, Loss: {'train/loss': 1.0580974817276, 'train/grad_norm_pre_clip': 3.475928783416748, 'timing/data': 0.0002681759651750326, 'timing/model': 1.2210356949362904, 'learning_rate/action_model': 9.620058304676928e-05, 'learning_rate/action_query_proj': 9.620058304676928e-05, 'learning_rate/base': 2.405014576169232e-05, 'epoch': 0.11}) +[2026-06-29 17:08:38,652][starVLA.training.train_starvla][INFO] - Step 256, Loss: {'train/loss': 0.27118536829948425, 'train/grad_norm_pre_clip': 6.910847187042236, 'timing/data': 0.00024510291405022144, 'timing/model': 1.2220678720623255, 'learning_rate/action_model': 9.617112867356178e-05, 'learning_rate/action_query_proj': 9.617112867356178e-05, 'learning_rate/base': 2.4042782168390445e-05, 'epoch': 0.11}) +[2026-06-29 17:09:11,004][starVLA.training.train_starvla][INFO] - Step 257, Loss: {'train/loss': 0.0007629175670444965, 'train/grad_norm_pre_clip': 5.7733588218688965, 'timing/data': 0.00019409717060625553, 'timing/model': 1.220252366969362, 'learning_rate/action_model': 9.614156531248522e-05, 'learning_rate/action_query_proj': 9.614156531248522e-05, 'learning_rate/base': 2.4035391328121305e-05, 'epoch': 0.11}) +[2026-06-29 17:09:43,355][starVLA.training.train_starvla][INFO] - Step 258, Loss: {'train/loss': 0.02759123407304287, 'train/grad_norm_pre_clip': 1.6928507089614868, 'timing/data': 0.00032766908407211304, 'timing/model': 1.22148113604635, 'learning_rate/action_model': 9.611189303648423e-05, 'learning_rate/action_query_proj': 9.611189303648423e-05, 'learning_rate/base': 2.4027973259121057e-05, 'epoch': 0.11}) +[2026-06-29 17:10:15,725][starVLA.training.train_starvla][INFO] - Step 259, Loss: {'train/loss': 0.009013419039547443, 'train/grad_norm_pre_clip': 2.523716926574707, 'timing/data': 0.0002515548840165138, 'timing/model': 1.2187766460701823, 'learning_rate/action_model': 9.60821119187722e-05, 'learning_rate/action_query_proj': 9.60821119187722e-05, 'learning_rate/base': 2.402052797969305e-05, 'epoch': 0.11}) +[2026-06-29 17:10:48,097][starVLA.training.train_starvla][INFO] - Step 260, Loss: {'train/loss': 0.08543598651885986, 'train/grad_norm_pre_clip': 1.624983549118042, 'timing/data': 0.0002009191084653139, 'timing/model': 1.222403883934021, 'learning_rate/action_model': 9.60522220328311e-05, 'learning_rate/action_query_proj': 9.60522220328311e-05, 'learning_rate/base': 2.4013055508207773e-05, 'epoch': 0.11}) +[2026-06-29 17:11:20,469][starVLA.training.train_starvla][INFO] - Step 261, Loss: {'train/loss': 0.35301005840301514, 'train/grad_norm_pre_clip': 6.368724346160889, 'timing/data': 0.0002416260540485382, 'timing/model': 1.2219574849586934, 'learning_rate/action_model': 9.602222345241122e-05, 'learning_rate/action_query_proj': 9.602222345241122e-05, 'learning_rate/base': 2.4005555863102805e-05, 'epoch': 0.11}) +[2026-06-29 17:11:52,851][starVLA.training.train_starvla][INFO] - Step 262, Loss: {'train/loss': 3.5850651329383254e-05, 'train/grad_norm_pre_clip': 4.542341232299805, 'timing/data': 0.00019625015556812286, 'timing/model': 1.2230847431346774, 'learning_rate/action_model': 9.59921162515311e-05, 'learning_rate/action_query_proj': 9.59921162515311e-05, 'learning_rate/base': 2.3998029062882776e-05, 'epoch': 0.11}) +[2026-06-29 17:12:25,234][starVLA.training.train_starvla][INFO] - Step 263, Loss: {'train/loss': 0.08092650026082993, 'train/grad_norm_pre_clip': 1.7074174880981445, 'timing/data': 0.00024799793027341366, 'timing/model': 1.2199544177856296, 'learning_rate/action_model': 9.596190050447726e-05, 'learning_rate/action_query_proj': 9.596190050447726e-05, 'learning_rate/base': 2.3990475126119315e-05, 'epoch': 0.11}) +[2026-06-29 17:12:57,587][starVLA.training.train_starvla][INFO] - Step 264, Loss: {'train/loss': 0.08357946574687958, 'train/grad_norm_pre_clip': 2.190314531326294, 'timing/data': 0.00019467808306217194, 'timing/model': 1.221439379034564, 'learning_rate/action_model': 9.593157628580404e-05, 'learning_rate/action_query_proj': 9.593157628580404e-05, 'learning_rate/base': 2.398289407145101e-05, 'epoch': 0.11}) +[2026-06-29 17:13:29,946][starVLA.training.train_starvla][INFO] - Step 265, Loss: {'train/loss': 0.018706753849983215, 'train/grad_norm_pre_clip': 2.1451644897460938, 'timing/data': 0.00019009900279343128, 'timing/model': 1.2203163711819798, 'learning_rate/action_model': 9.590114367033346e-05, 'learning_rate/action_query_proj': 9.590114367033346e-05, 'learning_rate/base': 2.3975285917583365e-05, 'epoch': 0.11}) +[2026-06-29 17:14:02,293][starVLA.training.train_starvla][INFO] - Step 266, Loss: {'train/loss': 0.09933796525001526, 'train/grad_norm_pre_clip': 3.721867561340332, 'timing/data': 0.0002641580067574978, 'timing/model': 1.2215893890243024, 'learning_rate/action_model': 9.587060273315493e-05, 'learning_rate/action_query_proj': 9.587060273315493e-05, 'learning_rate/base': 2.3967650683288733e-05, 'epoch': 0.11}) +[2026-06-29 17:14:34,663][starVLA.training.train_starvla][INFO] - Step 267, Loss: {'train/loss': 0.07526300847530365, 'train/grad_norm_pre_clip': 2.049278974533081, 'timing/data': 0.0002935039810836315, 'timing/model': 1.2200193649623543, 'learning_rate/action_model': 9.583995354962522e-05, 'learning_rate/action_query_proj': 9.583995354962522e-05, 'learning_rate/base': 2.3959988387406305e-05, 'epoch': 0.11}) +[2026-06-29 17:15:07,039][starVLA.training.train_starvla][INFO] - Step 268, Loss: {'train/loss': 0.014158991165459156, 'train/grad_norm_pre_clip': 3.176018714904785, 'timing/data': 0.00020815315656363964, 'timing/model': 1.2188440680038184, 'learning_rate/action_model': 9.580919619536814e-05, 'learning_rate/action_query_proj': 9.580919619536814e-05, 'learning_rate/base': 2.3952299048842036e-05, 'epoch': 0.12}) +[2026-06-29 17:15:39,412][starVLA.training.train_starvla][INFO] - Step 269, Loss: {'train/loss': 0.26339632272720337, 'train/grad_norm_pre_clip': 3.1615357398986816, 'timing/data': 0.0002078020479530096, 'timing/model': 1.224170396104455, 'learning_rate/action_model': 9.577833074627438e-05, 'learning_rate/action_query_proj': 9.577833074627438e-05, 'learning_rate/base': 2.3944582686568595e-05, 'epoch': 0.12}) +[2026-06-29 17:16:11,760][starVLA.training.train_starvla][INFO] - Step 270, Loss: {'train/loss': 0.0030489307828247547, 'train/grad_norm_pre_clip': 2.062844753265381, 'timing/data': 0.00019231298938393593, 'timing/model': 1.2176440458279103, 'learning_rate/action_model': 9.574735727850137e-05, 'learning_rate/action_query_proj': 9.574735727850137e-05, 'learning_rate/base': 2.3936839319625343e-05, 'epoch': 0.12}) +[2026-06-29 17:16:44,092][starVLA.training.train_starvla][INFO] - Step 271, Loss: {'train/loss': 0.001722955028526485, 'train/grad_norm_pre_clip': 2.6600451469421387, 'timing/data': 0.0001944480463862419, 'timing/model': 1.2196081229485571, 'learning_rate/action_model': 9.571627586847309e-05, 'learning_rate/action_query_proj': 9.571627586847309e-05, 'learning_rate/base': 2.3929068967118272e-05, 'epoch': 0.12}) +[2026-06-29 17:17:16,440][starVLA.training.train_starvla][INFO] - Step 272, Loss: {'train/loss': 2.8908211788802873e-06, 'train/grad_norm_pre_clip': 2.555968761444092, 'timing/data': 0.000363596947863698, 'timing/model': 1.2213714038953185, 'learning_rate/action_model': 9.568508659287979e-05, 'learning_rate/action_query_proj': 9.568508659287979e-05, 'learning_rate/base': 2.3921271648219947e-05, 'epoch': 0.12}) +[2026-06-29 17:17:48,771][starVLA.training.train_starvla][INFO] - Step 273, Loss: {'train/loss': 0.15544775128364563, 'train/grad_norm_pre_clip': 4.790022373199463, 'timing/data': 0.00019532907754182816, 'timing/model': 1.221321856137365, 'learning_rate/action_model': 9.565378952867794e-05, 'learning_rate/action_query_proj': 9.565378952867794e-05, 'learning_rate/base': 2.3913447382169484e-05, 'epoch': 0.12}) +[2026-06-29 17:18:21,119][starVLA.training.train_starvla][INFO] - Step 274, Loss: {'train/loss': 0.005785554181784391, 'train/grad_norm_pre_clip': 2.1716432571411133, 'timing/data': 0.00019363593310117722, 'timing/model': 1.2193266758695245, 'learning_rate/action_model': 9.562238475308993e-05, 'learning_rate/action_query_proj': 9.562238475308993e-05, 'learning_rate/base': 2.3905596188272482e-05, 'epoch': 0.12}) +[2026-06-29 17:18:53,490][starVLA.training.train_starvla][INFO] - Step 275, Loss: {'train/loss': 0.003679160261526704, 'train/grad_norm_pre_clip': 2.8920400142669678, 'timing/data': 0.00025142496451735497, 'timing/model': 1.2226012360770255, 'learning_rate/action_model': 9.559087234360391e-05, 'learning_rate/action_query_proj': 9.559087234360391e-05, 'learning_rate/base': 2.3897718085900978e-05, 'epoch': 0.12}) +[2026-06-29 17:19:25,836][starVLA.training.train_starvla][INFO] - Step 276, Loss: {'train/loss': 1.4364512026077136e-05, 'train/grad_norm_pre_clip': 1.9696069955825806, 'timing/data': 0.0001945970579981804, 'timing/model': 1.2212154110893607, 'learning_rate/action_model': 9.555925237797362e-05, 'learning_rate/action_query_proj': 9.555925237797362e-05, 'learning_rate/base': 2.3889813094493405e-05, 'epoch': 0.12}) +[2026-06-29 17:19:58,157][starVLA.training.train_starvla][INFO] - Step 277, Loss: {'train/loss': 0.2090844362974167, 'train/grad_norm_pre_clip': 5.079381465911865, 'timing/data': 0.00019590998999774456, 'timing/model': 1.2201428499538451, 'learning_rate/action_model': 9.55275249342182e-05, 'learning_rate/action_query_proj': 9.55275249342182e-05, 'learning_rate/base': 2.388188123355455e-05, 'epoch': 0.12}) +[2026-06-29 17:20:30,503][starVLA.training.train_starvla][INFO] - Step 278, Loss: {'train/loss': 0.02635415457189083, 'train/grad_norm_pre_clip': 5.185588836669922, 'timing/data': 0.00022525503300130367, 'timing/model': 1.2230245529208332, 'learning_rate/action_model': 9.549569009062191e-05, 'learning_rate/action_query_proj': 9.549569009062191e-05, 'learning_rate/base': 2.387392252265548e-05, 'epoch': 0.12}) +[2026-06-29 17:21:02,867][starVLA.training.train_starvla][INFO] - Step 279, Loss: {'train/loss': 0.016640206798911095, 'train/grad_norm_pre_clip': 2.7791666984558105, 'timing/data': 0.0002666129730641842, 'timing/model': 1.2196692971047014, 'learning_rate/action_model': 9.546374792573414e-05, 'learning_rate/action_query_proj': 9.546374792573414e-05, 'learning_rate/base': 2.3865936981433535e-05, 'epoch': 0.12}) +[2026-06-29 17:21:35,230][starVLA.training.train_starvla][INFO] - Step 280, Loss: {'train/loss': 0.2343670129776001, 'train/grad_norm_pre_clip': 2.1207048892974854, 'timing/data': 0.00019381497986614704, 'timing/model': 1.223613176960498, 'learning_rate/action_model': 9.543169851836894e-05, 'learning_rate/action_query_proj': 9.543169851836894e-05, 'learning_rate/base': 2.3857924629592234e-05, 'epoch': 0.12}) +[2026-06-29 17:22:07,594][starVLA.training.train_starvla][INFO] - Step 281, Loss: {'train/loss': 0.3372442424297333, 'train/grad_norm_pre_clip': 2.2240750789642334, 'timing/data': 0.00021139904856681824, 'timing/model': 1.2228309400379658, 'learning_rate/action_model': 9.539954194760509e-05, 'learning_rate/action_query_proj': 9.539954194760509e-05, 'learning_rate/base': 2.3849885486901273e-05, 'epoch': 0.12}) +[2026-06-29 17:22:39,942][starVLA.training.train_starvla][INFO] - Step 282, Loss: {'train/loss': 0.43801841139793396, 'train/grad_norm_pre_clip': 3.538799524307251, 'timing/data': 0.0002472260966897011, 'timing/model': 1.2173277740366757, 'learning_rate/action_model': 9.536727829278568e-05, 'learning_rate/action_query_proj': 9.536727829278568e-05, 'learning_rate/base': 2.384181957319642e-05, 'epoch': 0.12}) +[2026-06-29 17:23:12,292][starVLA.training.train_starvla][INFO] - Step 283, Loss: {'train/loss': 0.10237559676170349, 'train/grad_norm_pre_clip': 2.9520530700683594, 'timing/data': 0.000247396994382143, 'timing/model': 1.22043427801691, 'learning_rate/action_model': 9.533490763351811e-05, 'learning_rate/action_query_proj': 9.533490763351811e-05, 'learning_rate/base': 2.383372690837953e-05, 'epoch': 0.12}) +[2026-06-29 17:23:44,649][starVLA.training.train_starvla][INFO] - Step 284, Loss: {'train/loss': 0.06512557715177536, 'train/grad_norm_pre_clip': 2.7262163162231445, 'timing/data': 0.00019712187349796295, 'timing/model': 1.2232560459524393, 'learning_rate/action_model': 9.530243004967375e-05, 'learning_rate/action_query_proj': 9.530243004967375e-05, 'learning_rate/base': 2.3825607512418436e-05, 'epoch': 0.12}) +[2026-06-29 17:24:17,039][starVLA.training.train_starvla][INFO] - Step 285, Loss: {'train/loss': 0.019227685406804085, 'train/grad_norm_pre_clip': 3.9509596824645996, 'timing/data': 0.0003457521088421345, 'timing/model': 1.2206241919193417, 'learning_rate/action_model': 9.526984562138779e-05, 'learning_rate/action_query_proj': 9.526984562138779e-05, 'learning_rate/base': 2.3817461405346946e-05, 'epoch': 0.12}) +[2026-06-29 17:24:49,399][starVLA.training.train_starvla][INFO] - Step 286, Loss: {'train/loss': 0.03694091737270355, 'train/grad_norm_pre_clip': 1.9941381216049194, 'timing/data': 0.00019321381114423275, 'timing/model': 1.2193711220752448, 'learning_rate/action_model': 9.523715442905912e-05, 'learning_rate/action_query_proj': 9.523715442905912e-05, 'learning_rate/base': 2.380928860726478e-05, 'epoch': 0.12}) +[2026-06-29 17:25:21,735][starVLA.training.train_starvla][INFO] - Step 287, Loss: {'train/loss': 0.2725929319858551, 'train/grad_norm_pre_clip': 6.959357738494873, 'timing/data': 0.0001891071442514658, 'timing/model': 1.219405923038721, 'learning_rate/action_model': 9.520435655334995e-05, 'learning_rate/action_query_proj': 9.520435655334995e-05, 'learning_rate/base': 2.3801089138337487e-05, 'epoch': 0.12}) +[2026-06-29 17:25:54,085][starVLA.training.train_starvla][INFO] - Step 288, Loss: {'train/loss': 0.3992690443992615, 'train/grad_norm_pre_clip': 5.707257270812988, 'timing/data': 0.00020420504733920097, 'timing/model': 1.2229637638665736, 'learning_rate/action_model': 9.517145207518581e-05, 'learning_rate/action_query_proj': 9.517145207518581e-05, 'learning_rate/base': 2.3792863018796453e-05, 'epoch': 0.12}) +[2026-06-29 17:26:26,609][starVLA.training.train_starvla][INFO] - Step 289, Loss: {'train/loss': 0.03107585944235325, 'train/grad_norm_pre_clip': 4.026444911956787, 'timing/data': 0.0002515050582587719, 'timing/model': 1.2221094688866287, 'learning_rate/action_model': 9.513844107575523e-05, 'learning_rate/action_query_proj': 9.513844107575523e-05, 'learning_rate/base': 2.378461026893881e-05, 'epoch': 0.12}) +[2026-06-29 17:26:58,985][starVLA.training.train_starvla][INFO] - Step 290, Loss: {'train/loss': 0.011574739590287209, 'train/grad_norm_pre_clip': 2.396954298019409, 'timing/data': 0.00020692008547484875, 'timing/model': 1.2204002710059285, 'learning_rate/action_model': 9.510532363650955e-05, 'learning_rate/action_query_proj': 9.510532363650955e-05, 'learning_rate/base': 2.3776330909127388e-05, 'epoch': 0.12}) +[2026-06-29 17:27:31,371][starVLA.training.train_starvla][INFO] - Step 291, Loss: {'train/loss': 0.25309187173843384, 'train/grad_norm_pre_clip': 1.85785973072052, 'timing/data': 0.0002551809884607792, 'timing/model': 1.2200681939721107, 'learning_rate/action_model': 9.507209983916279e-05, 'learning_rate/action_query_proj': 9.507209983916279e-05, 'learning_rate/base': 2.3768024959790697e-05, 'epoch': 0.13}) +[2026-06-29 17:28:03,756][starVLA.training.train_starvla][INFO] - Step 292, Loss: {'train/loss': 0.09010729193687439, 'train/grad_norm_pre_clip': 6.040132999420166, 'timing/data': 0.00022996379993855953, 'timing/model': 1.2128161888103932, 'learning_rate/action_model': 9.503876976569133e-05, 'learning_rate/action_query_proj': 9.503876976569133e-05, 'learning_rate/base': 2.3759692441422832e-05, 'epoch': 0.13}) +[2026-06-29 17:28:36,104][starVLA.training.train_starvla][INFO] - Step 293, Loss: {'train/loss': 0.20150752365589142, 'train/grad_norm_pre_clip': 2.862492561340332, 'timing/data': 0.0001982739195227623, 'timing/model': 1.2214863090775907, 'learning_rate/action_model': 9.500533349833382e-05, 'learning_rate/action_query_proj': 9.500533349833382e-05, 'learning_rate/base': 2.3751333374583456e-05, 'epoch': 0.13}) +[2026-06-29 17:29:08,372][starVLA.training.train_starvla][INFO] - Step 294, Loss: {'train/loss': 0.16714458167552948, 'train/grad_norm_pre_clip': 4.8180413246154785, 'timing/data': 0.00020025810226798058, 'timing/model': 1.2188766659237444, 'learning_rate/action_model': 9.497179111959098e-05, 'learning_rate/action_query_proj': 9.497179111959098e-05, 'learning_rate/base': 2.3742947779897744e-05, 'epoch': 0.13}) +[2026-06-29 17:29:40,713][starVLA.training.train_starvla][INFO] - Step 295, Loss: {'train/loss': 0.02812219224870205, 'train/grad_norm_pre_clip': 4.343989372253418, 'timing/data': 0.0002504219301044941, 'timing/model': 1.2242993488907814, 'learning_rate/action_model': 9.493814271222521e-05, 'learning_rate/action_query_proj': 9.493814271222521e-05, 'learning_rate/base': 2.3734535678056302e-05, 'epoch': 0.13}) +[2026-06-29 17:30:13,044][starVLA.training.train_starvla][INFO] - Step 296, Loss: {'train/loss': 0.03390532359480858, 'train/grad_norm_pre_clip': 3.914971113204956, 'timing/data': 0.00019567878916859627, 'timing/model': 1.221466857008636, 'learning_rate/action_model': 9.490438835926067e-05, 'learning_rate/action_query_proj': 9.490438835926067e-05, 'learning_rate/base': 2.3726097089815167e-05, 'epoch': 0.13}) +[2026-06-29 17:30:45,408][starVLA.training.train_starvla][INFO] - Step 297, Loss: {'train/loss': 0.044021960347890854, 'train/grad_norm_pre_clip': 3.4569313526153564, 'timing/data': 0.00041719782166182995, 'timing/model': 1.218708995031193, 'learning_rate/action_model': 9.487052814398284e-05, 'learning_rate/action_query_proj': 9.487052814398284e-05, 'learning_rate/base': 2.371763203599571e-05, 'epoch': 0.13}) +[2026-06-29 17:31:17,728][starVLA.training.train_starvla][INFO] - Step 298, Loss: {'train/loss': 0.023662813007831573, 'train/grad_norm_pre_clip': 1.6434589624404907, 'timing/data': 0.0002129219938069582, 'timing/model': 1.2241456771735102, 'learning_rate/action_model': 9.483656214993847e-05, 'learning_rate/action_query_proj': 9.483656214993847e-05, 'learning_rate/base': 2.3709140537484617e-05, 'epoch': 0.13}) +[2026-06-29 17:31:50,080][starVLA.training.train_starvla][INFO] - Step 299, Loss: {'train/loss': 2.95042423203995e-06, 'train/grad_norm_pre_clip': 2.1346311569213867, 'timing/data': 0.00022992398589849472, 'timing/model': 1.2231899360194802, 'learning_rate/action_model': 9.480249046093523e-05, 'learning_rate/action_query_proj': 9.480249046093523e-05, 'learning_rate/base': 2.3700622615233807e-05, 'epoch': 0.13}) +[2026-06-29 17:32:22,426][starVLA.training.train_starvla][INFO] - Step 300, Loss: {'train/loss': 0.0006136730662547052, 'train/grad_norm_pre_clip': 6.11156702041626, 'timing/data': 0.00018905592150986195, 'timing/model': 1.2195981219410896, 'learning_rate/action_model': 9.476831316104165e-05, 'learning_rate/action_query_proj': 9.476831316104165e-05, 'learning_rate/base': 2.3692078290260413e-05, 'epoch': 0.13}) +[2026-06-29 17:32:54,785][starVLA.training.train_starvla][INFO] - Step 301, Loss: {'train/loss': 0.16526754200458527, 'train/grad_norm_pre_clip': 4.038857460021973, 'timing/data': 0.0002989440690726042, 'timing/model': 1.2230872309301049, 'learning_rate/action_model': 9.473403033458684e-05, 'learning_rate/action_query_proj': 9.473403033458684e-05, 'learning_rate/base': 2.368350758364671e-05, 'epoch': 0.13}) +[2026-06-29 17:33:27,174][starVLA.training.train_starvla][INFO] - Step 302, Loss: {'train/loss': 0.23701812326908112, 'train/grad_norm_pre_clip': 4.444374084472656, 'timing/data': 0.00028318515978753567, 'timing/model': 1.2209516048897058, 'learning_rate/action_model': 9.469964206616023e-05, 'learning_rate/action_query_proj': 9.469964206616023e-05, 'learning_rate/base': 2.3674910516540058e-05, 'epoch': 0.13}) +[2026-06-29 17:33:59,529][starVLA.training.train_starvla][INFO] - Step 303, Loss: {'train/loss': 0.04567444697022438, 'train/grad_norm_pre_clip': 2.450347661972046, 'timing/data': 0.0003369450569152832, 'timing/model': 1.222099517006427, 'learning_rate/action_model': 9.466514844061146e-05, 'learning_rate/action_query_proj': 9.466514844061146e-05, 'learning_rate/base': 2.3666287110152866e-05, 'epoch': 0.13}) +[2026-06-29 17:34:31,899][starVLA.training.train_starvla][INFO] - Step 304, Loss: {'train/loss': 0.41491344571113586, 'train/grad_norm_pre_clip': 2.688110828399658, 'timing/data': 0.0003466331399977207, 'timing/model': 1.219768397975713, 'learning_rate/action_model': 9.463054954305014e-05, 'learning_rate/action_query_proj': 9.463054954305014e-05, 'learning_rate/base': 2.3657637385762534e-05, 'epoch': 0.13}) +[2026-06-29 17:35:04,242][starVLA.training.train_starvla][INFO] - Step 305, Loss: {'train/loss': 0.04742182046175003, 'train/grad_norm_pre_clip': 4.088337421417236, 'timing/data': 0.0002412551548331976, 'timing/model': 1.220878018066287, 'learning_rate/action_model': 9.459584545884562e-05, 'learning_rate/action_query_proj': 9.459584545884562e-05, 'learning_rate/base': 2.3648961364711404e-05, 'epoch': 0.13}) +[2026-06-29 17:35:36,576][starVLA.training.train_starvla][INFO] - Step 306, Loss: {'train/loss': 0.006329897791147232, 'train/grad_norm_pre_clip': 4.574312210083008, 'timing/data': 0.00020102900452911854, 'timing/model': 1.2245559620205313, 'learning_rate/action_model': 9.456103627362673e-05, 'learning_rate/action_query_proj': 9.456103627362673e-05, 'learning_rate/base': 2.364025906840668e-05, 'epoch': 0.13}) +[2026-06-29 17:36:08,935][starVLA.training.train_starvla][INFO] - Step 307, Loss: {'train/loss': 0.06680146604776382, 'train/grad_norm_pre_clip': 3.8009748458862305, 'timing/data': 0.0002480179537087679, 'timing/model': 1.220284892944619, 'learning_rate/action_model': 9.452612207328168e-05, 'learning_rate/action_query_proj': 9.452612207328168e-05, 'learning_rate/base': 2.363153051832042e-05, 'epoch': 0.13}) +[2026-06-29 17:36:41,261][starVLA.training.train_starvla][INFO] - Step 308, Loss: {'train/loss': 0.058051642030477524, 'train/grad_norm_pre_clip': 7.030772686004639, 'timing/data': 0.00019169203005731106, 'timing/model': 1.2200784410815686, 'learning_rate/action_model': 9.449110294395786e-05, 'learning_rate/action_query_proj': 9.449110294395786e-05, 'learning_rate/base': 2.3622775735989464e-05, 'epoch': 0.13}) +[2026-06-29 17:37:13,644][starVLA.training.train_starvla][INFO] - Step 309, Loss: {'train/loss': 0.08793892711400986, 'train/grad_norm_pre_clip': 1.954013705253601, 'timing/data': 0.00020459596998989582, 'timing/model': 1.2220256139989942, 'learning_rate/action_model': 9.445597897206142e-05, 'learning_rate/action_query_proj': 9.445597897206142e-05, 'learning_rate/base': 2.3613994743015355e-05, 'epoch': 0.13}) +[2026-06-29 17:37:46,010][starVLA.training.train_starvla][INFO] - Step 310, Loss: {'train/loss': 0.003753179684281349, 'train/grad_norm_pre_clip': 1.4383842945098877, 'timing/data': 0.0002609621733427048, 'timing/model': 1.2256998380180448, 'learning_rate/action_model': 9.44207502442573e-05, 'learning_rate/action_query_proj': 9.44207502442573e-05, 'learning_rate/base': 2.3605187561064324e-05, 'epoch': 0.13}) +[2026-06-29 17:38:18,344][starVLA.training.train_starvla][INFO] - Step 311, Loss: {'train/loss': 0.031912654638290405, 'train/grad_norm_pre_clip': 2.011695146560669, 'timing/data': 0.00020296196453273296, 'timing/model': 1.22346095694229, 'learning_rate/action_model': 9.438541684746885e-05, 'learning_rate/action_query_proj': 9.438541684746885e-05, 'learning_rate/base': 2.3596354211867213e-05, 'epoch': 0.13}) +[2026-06-29 17:38:50,606][starVLA.training.train_starvla][INFO] - Step 312, Loss: {'train/loss': 0.15844391286373138, 'train/grad_norm_pre_clip': 6.055861473083496, 'timing/data': 0.00019431603141129017, 'timing/model': 1.2216445470694453, 'learning_rate/action_model': 9.434997886887776e-05, 'learning_rate/action_query_proj': 9.434997886887776e-05, 'learning_rate/base': 2.358749471721944e-05, 'epoch': 0.13}) +[2026-06-29 17:39:22,945][starVLA.training.train_starvla][INFO] - Step 313, Loss: {'train/loss': 0.0008293753489851952, 'train/grad_norm_pre_clip': 2.7904858589172363, 'timing/data': 0.0003004958853125572, 'timing/model': 1.2194474849384278, 'learning_rate/action_model': 9.431443639592369e-05, 'learning_rate/action_query_proj': 9.431443639592369e-05, 'learning_rate/base': 2.3578609098980922e-05, 'epoch': 0.13}) +[2026-06-29 17:39:55,303][starVLA.training.train_starvla][INFO] - Step 314, Loss: {'train/loss': 0.06364797800779343, 'train/grad_norm_pre_clip': 5.0234198570251465, 'timing/data': 0.00019088992848992348, 'timing/model': 1.221473332028836, 'learning_rate/action_model': 9.427878951630418e-05, 'learning_rate/action_query_proj': 9.427878951630418e-05, 'learning_rate/base': 2.3569697379076044e-05, 'epoch': 0.14}) +[2026-06-29 17:40:27,645][starVLA.training.train_starvla][INFO] - Step 315, Loss: {'train/loss': 0.09824433922767639, 'train/grad_norm_pre_clip': 1.824880838394165, 'timing/data': 0.00018951809033751488, 'timing/model': 1.2168781580403447, 'learning_rate/action_model': 9.424303831797433e-05, 'learning_rate/action_query_proj': 9.424303831797433e-05, 'learning_rate/base': 2.3560759579493584e-05, 'epoch': 0.14}) +[2026-06-29 17:40:59,950][starVLA.training.train_starvla][INFO] - Step 316, Loss: {'train/loss': 0.30374255776405334, 'train/grad_norm_pre_clip': 1.7877167463302612, 'timing/data': 0.00019570998847484589, 'timing/model': 1.2196525691542774, 'learning_rate/action_model': 9.42071828891467e-05, 'learning_rate/action_query_proj': 9.42071828891467e-05, 'learning_rate/base': 2.3551795722286675e-05, 'epoch': 0.14}) +[2026-06-29 17:41:32,369][starVLA.training.train_starvla][INFO] - Step 317, Loss: {'train/loss': 0.004292003344744444, 'train/grad_norm_pre_clip': 1.7131229639053345, 'timing/data': 0.00019583012908697128, 'timing/model': 1.2206287649460137, 'learning_rate/action_model': 9.417122331829097e-05, 'learning_rate/action_query_proj': 9.417122331829097e-05, 'learning_rate/base': 2.3542805829572743e-05, 'epoch': 0.14}) +[2026-06-29 17:42:04,714][starVLA.training.train_starvla][INFO] - Step 318, Loss: {'train/loss': 0.18902979791164398, 'train/grad_norm_pre_clip': 2.6974265575408936, 'timing/data': 0.0002957370597869158, 'timing/model': 1.221885577077046, 'learning_rate/action_model': 9.413515969413386e-05, 'learning_rate/action_query_proj': 9.413515969413386e-05, 'learning_rate/base': 2.3533789923533465e-05, 'epoch': 0.14}) +[2026-06-29 17:42:37,037][starVLA.training.train_starvla][INFO] - Step 319, Loss: {'train/loss': 0.015805523842573166, 'train/grad_norm_pre_clip': 3.0212786197662354, 'timing/data': 0.00020005716942250729, 'timing/model': 1.2193058200646192, 'learning_rate/action_model': 9.40989921056587e-05, 'learning_rate/action_query_proj': 9.40989921056587e-05, 'learning_rate/base': 2.3524748026414675e-05, 'epoch': 0.14}) +[2026-06-29 17:43:09,391][starVLA.training.train_starvla][INFO] - Step 320, Loss: {'train/loss': 0.158325657248497, 'train/grad_norm_pre_clip': 3.352287530899048, 'timing/data': 0.0002140540163964033, 'timing/model': 1.2206808300688863, 'learning_rate/action_model': 9.406272064210545e-05, 'learning_rate/action_query_proj': 9.406272064210545e-05, 'learning_rate/base': 2.351568016052636e-05, 'epoch': 0.14}) +[2026-06-29 17:43:41,710][starVLA.training.train_starvla][INFO] - Step 321, Loss: {'train/loss': 0.29407650232315063, 'train/grad_norm_pre_clip': 2.3411333560943604, 'timing/data': 0.00020106020383536816, 'timing/model': 1.221269462024793, 'learning_rate/action_model': 9.402634539297037e-05, 'learning_rate/action_query_proj': 9.402634539297037e-05, 'learning_rate/base': 2.350658634824259e-05, 'epoch': 0.14}) +[2026-06-29 17:44:14,043][starVLA.training.train_starvla][INFO] - Step 322, Loss: {'train/loss': 0.09491155296564102, 'train/grad_norm_pre_clip': 2.051621913909912, 'timing/data': 0.0002543400041759014, 'timing/model': 1.223741594934836, 'learning_rate/action_model': 9.398986644800573e-05, 'learning_rate/action_query_proj': 9.398986644800573e-05, 'learning_rate/base': 2.3497466612001434e-05, 'epoch': 0.14}) +[2026-06-29 17:44:46,390][starVLA.training.train_starvla][INFO] - Step 323, Loss: {'train/loss': 0.840561032295227, 'train/grad_norm_pre_clip': 4.158320903778076, 'timing/data': 0.0002020108513534069, 'timing/model': 1.2196722279768437, 'learning_rate/action_model': 9.395328389721974e-05, 'learning_rate/action_query_proj': 9.395328389721974e-05, 'learning_rate/base': 2.3488320974304935e-05, 'epoch': 0.14}) +[2026-06-29 17:45:18,766][starVLA.training.train_starvla][INFO] - Step 324, Loss: {'train/loss': 0.0021239446941763163, 'train/grad_norm_pre_clip': 4.4979424476623535, 'timing/data': 0.000309955095872283, 'timing/model': 1.221251280978322, 'learning_rate/action_model': 9.391659783087618e-05, 'learning_rate/action_query_proj': 9.391659783087618e-05, 'learning_rate/base': 2.3479149457719045e-05, 'epoch': 0.14}) +[2026-06-29 17:45:51,128][starVLA.training.train_starvla][INFO] - Step 325, Loss: {'train/loss': 0.011881011538207531, 'train/grad_norm_pre_clip': 2.534682512283325, 'timing/data': 0.00025884900242090225, 'timing/model': 1.222152424044907, 'learning_rate/action_model': 9.387980833949427e-05, 'learning_rate/action_query_proj': 9.387980833949427e-05, 'learning_rate/base': 2.3469952084873568e-05, 'epoch': 0.14}) +[2026-06-29 17:46:23,467][starVLA.training.train_starvla][INFO] - Step 326, Loss: {'train/loss': 0.17697204649448395, 'train/grad_norm_pre_clip': 3.9602506160736084, 'timing/data': 0.00025639403611421585, 'timing/model': 1.2214906269218773, 'learning_rate/action_model': 9.384291551384843e-05, 'learning_rate/action_query_proj': 9.384291551384843e-05, 'learning_rate/base': 2.346072887846211e-05, 'epoch': 0.14}) +[2026-06-29 17:46:55,793][starVLA.training.train_starvla][INFO] - Step 327, Loss: {'train/loss': 0.00262161111459136, 'train/grad_norm_pre_clip': 1.9449021816253662, 'timing/data': 0.00019567995332181454, 'timing/model': 1.219718883978203, 'learning_rate/action_model': 9.380591944496805e-05, 'learning_rate/action_query_proj': 9.380591944496805e-05, 'learning_rate/base': 2.3451479861242013e-05, 'epoch': 0.14}) +[2026-06-29 17:47:28,257][starVLA.training.train_starvla][INFO] - Step 328, Loss: {'train/loss': 0.13389886915683746, 'train/grad_norm_pre_clip': 2.5922796726226807, 'timing/data': 0.00023963209241628647, 'timing/model': 1.2224745298735797, 'learning_rate/action_model': 9.376882022413723e-05, 'learning_rate/action_query_proj': 9.376882022413723e-05, 'learning_rate/base': 2.3442205056034307e-05, 'epoch': 0.14}) +[2026-06-29 17:48:00,587][starVLA.training.train_starvla][INFO] - Step 329, Loss: {'train/loss': 0.06209295988082886, 'train/grad_norm_pre_clip': 5.669386863708496, 'timing/data': 0.00024997093714773655, 'timing/model': 1.2185800599399954, 'learning_rate/action_model': 9.373161794289465e-05, 'learning_rate/action_query_proj': 9.373161794289465e-05, 'learning_rate/base': 2.3432904485723662e-05, 'epoch': 0.14}) +[2026-06-29 17:48:32,915][starVLA.training.train_starvla][INFO] - Step 330, Loss: {'train/loss': 0.21748310327529907, 'train/grad_norm_pre_clip': 5.104102611541748, 'timing/data': 0.0002933931536972523, 'timing/model': 1.2208391050808132, 'learning_rate/action_model': 9.36943126930332e-05, 'learning_rate/action_query_proj': 9.36943126930332e-05, 'learning_rate/base': 2.34235781732583e-05, 'epoch': 0.14}) +[2026-06-29 17:49:05,242][starVLA.training.train_starvla][INFO] - Step 331, Loss: {'train/loss': 0.07786156982183456, 'train/grad_norm_pre_clip': 5.8882832527160645, 'timing/data': 0.0001960298977792263, 'timing/model': 1.220264834817499, 'learning_rate/action_model': 9.365690456659987e-05, 'learning_rate/action_query_proj': 9.365690456659987e-05, 'learning_rate/base': 2.3414226141649968e-05, 'epoch': 0.14}) +[2026-06-29 17:49:37,574][starVLA.training.train_starvla][INFO] - Step 332, Loss: {'train/loss': 0.5321996212005615, 'train/grad_norm_pre_clip': 5.709505081176758, 'timing/data': 0.0002454030327498913, 'timing/model': 1.219172626035288, 'learning_rate/action_model': 9.361939365589553e-05, 'learning_rate/action_query_proj': 9.361939365589553e-05, 'learning_rate/base': 2.3404848413973883e-05, 'epoch': 0.14}) +[2026-06-29 17:50:09,894][starVLA.training.train_starvla][INFO] - Step 333, Loss: {'train/loss': 0.11862780153751373, 'train/grad_norm_pre_clip': 2.214237928390503, 'timing/data': 0.0001927840057760477, 'timing/model': 1.2246069971006364, 'learning_rate/action_model': 9.358178005347461e-05, 'learning_rate/action_query_proj': 9.358178005347461e-05, 'learning_rate/base': 2.3395445013368653e-05, 'epoch': 0.14}) +[2026-06-29 17:50:42,227][starVLA.training.train_starvla][INFO] - Step 334, Loss: {'train/loss': 0.04124177247285843, 'train/grad_norm_pre_clip': 2.412841796875, 'timing/data': 0.00032055494375526905, 'timing/model': 1.2223275860305876, 'learning_rate/action_model': 9.354406385214491e-05, 'learning_rate/action_query_proj': 9.354406385214491e-05, 'learning_rate/base': 2.3386015963036227e-05, 'epoch': 0.14}) +[2026-06-29 17:51:14,553][starVLA.training.train_starvla][INFO] - Step 335, Loss: {'train/loss': 0.0017191721126437187, 'train/grad_norm_pre_clip': 6.213620185852051, 'timing/data': 0.0001987649593502283, 'timing/model': 1.2230184059590101, 'learning_rate/action_model': 9.350624514496746e-05, 'learning_rate/action_query_proj': 9.350624514496746e-05, 'learning_rate/base': 2.3376561286241865e-05, 'epoch': 0.14}) +[2026-06-29 17:51:46,877][starVLA.training.train_starvla][INFO] - Step 336, Loss: {'train/loss': 0.21315008401870728, 'train/grad_norm_pre_clip': 4.042799949645996, 'timing/data': 0.0002442409750074148, 'timing/model': 1.2216531280428171, 'learning_rate/action_model': 9.34683240252561e-05, 'learning_rate/action_query_proj': 9.34683240252561e-05, 'learning_rate/base': 2.3367081006314024e-05, 'epoch': 0.14}) +[2026-06-29 17:52:19,212][starVLA.training.train_starvla][INFO] - Step 337, Loss: {'train/loss': 0.29104650020599365, 'train/grad_norm_pre_clip': 3.9549107551574707, 'timing/data': 0.00024409987963736057, 'timing/model': 1.2198366757947952, 'learning_rate/action_model': 9.343030058657747e-05, 'learning_rate/action_query_proj': 9.343030058657747e-05, 'learning_rate/base': 2.3357575146644368e-05, 'epoch': 0.15}) +[2026-06-29 17:52:51,570][starVLA.training.train_starvla][INFO] - Step 338, Loss: {'train/loss': 0.07343820482492447, 'train/grad_norm_pre_clip': 2.4005444049835205, 'timing/data': 0.00024497206322848797, 'timing/model': 1.2189987301826477, 'learning_rate/action_model': 9.33921749227506e-05, 'learning_rate/action_query_proj': 9.33921749227506e-05, 'learning_rate/base': 2.334804373068765e-05, 'epoch': 0.15}) +[2026-06-29 17:53:23,950][starVLA.training.train_starvla][INFO] - Step 339, Loss: {'train/loss': 0.10650745779275894, 'train/grad_norm_pre_clip': 3.1033682823181152, 'timing/data': 0.0001876938622444868, 'timing/model': 1.2214805248659104, 'learning_rate/action_model': 9.335394712784677e-05, 'learning_rate/action_query_proj': 9.335394712784677e-05, 'learning_rate/base': 2.3338486781961693e-05, 'epoch': 0.15}) +[2026-06-29 17:53:56,294][starVLA.training.train_starvla][INFO] - Step 340, Loss: {'train/loss': 0.13661789894104004, 'train/grad_norm_pre_clip': 1.2421815395355225, 'timing/data': 0.00024166610091924667, 'timing/model': 1.2203456468414515, 'learning_rate/action_model': 9.33156172961893e-05, 'learning_rate/action_query_proj': 9.33156172961893e-05, 'learning_rate/base': 2.3328904324047325e-05, 'epoch': 0.15}) +[2026-06-29 17:54:28,626][starVLA.training.train_starvla][INFO] - Step 341, Loss: {'train/loss': 0.11363265663385391, 'train/grad_norm_pre_clip': 5.553390979766846, 'timing/data': 0.000245923874899745, 'timing/model': 1.2228833280969411, 'learning_rate/action_model': 9.327718552235321e-05, 'learning_rate/action_query_proj': 9.327718552235321e-05, 'learning_rate/base': 2.3319296380588303e-05, 'epoch': 0.15}) +[2026-06-29 17:55:00,955][starVLA.training.train_starvla][INFO] - Step 342, Loss: {'train/loss': 0.015981843695044518, 'train/grad_norm_pre_clip': 5.630172252655029, 'timing/data': 0.00019849510863423347, 'timing/model': 1.2215872439555824, 'learning_rate/action_model': 9.323865190116508e-05, 'learning_rate/action_query_proj': 9.323865190116508e-05, 'learning_rate/base': 2.330966297529127e-05, 'epoch': 0.15}) +[2026-06-29 17:55:33,271][starVLA.training.train_starvla][INFO] - Step 343, Loss: {'train/loss': 0.07386057078838348, 'train/grad_norm_pre_clip': 7.601175308227539, 'timing/data': 0.00028685200959444046, 'timing/model': 1.211281580850482, 'learning_rate/action_model': 9.320001652770279e-05, 'learning_rate/action_query_proj': 9.320001652770279e-05, 'learning_rate/base': 2.3300004131925698e-05, 'epoch': 0.15}) +[2026-06-29 17:56:05,624][starVLA.training.train_starvla][INFO] - Step 344, Loss: {'train/loss': 0.12850792706012726, 'train/grad_norm_pre_clip': 3.5360665321350098, 'timing/data': 0.00030726008117198944, 'timing/model': 1.2207906311377883, 'learning_rate/action_model': 9.316127949729531e-05, 'learning_rate/action_query_proj': 9.316127949729531e-05, 'learning_rate/base': 2.329031987432383e-05, 'epoch': 0.15}) +[2026-06-29 17:56:37,951][starVLA.training.train_starvla][INFO] - Step 345, Loss: {'train/loss': 1.0918946266174316, 'train/grad_norm_pre_clip': 3.317121744155884, 'timing/data': 0.00027711293660104275, 'timing/model': 1.2189850041177124, 'learning_rate/action_model': 9.312244090552239e-05, 'learning_rate/action_query_proj': 9.312244090552239e-05, 'learning_rate/base': 2.3280610226380596e-05, 'epoch': 0.15}) +[2026-06-29 17:57:10,264][starVLA.training.train_starvla][INFO] - Step 346, Loss: {'train/loss': 0.08340675383806229, 'train/grad_norm_pre_clip': 1.2918462753295898, 'timing/data': 0.0003062079194933176, 'timing/model': 1.2183247569482774, 'learning_rate/action_model': 9.308350084821439e-05, 'learning_rate/action_query_proj': 9.308350084821439e-05, 'learning_rate/base': 2.3270875212053597e-05, 'epoch': 0.15}) +[2026-06-29 17:57:42,580][starVLA.training.train_starvla][INFO] - Step 347, Loss: {'train/loss': 0.14700528979301453, 'train/grad_norm_pre_clip': 2.3540589809417725, 'timing/data': 0.00026233517564833164, 'timing/model': 1.219234460964799, 'learning_rate/action_model': 9.304445942145204e-05, 'learning_rate/action_query_proj': 9.304445942145204e-05, 'learning_rate/base': 2.326111485536301e-05, 'epoch': 0.15}) +[2026-06-29 17:58:14,956][starVLA.training.train_starvla][INFO] - Step 348, Loss: {'train/loss': 0.05776374787092209, 'train/grad_norm_pre_clip': 3.794894218444824, 'timing/data': 0.0003741360269486904, 'timing/model': 1.223021614132449, 'learning_rate/action_model': 9.300531672156618e-05, 'learning_rate/action_query_proj': 9.300531672156618e-05, 'learning_rate/base': 2.3251329180391544e-05, 'epoch': 0.15}) +[2026-06-29 17:58:47,328][starVLA.training.train_starvla][INFO] - Step 349, Loss: {'train/loss': 0.007832065224647522, 'train/grad_norm_pre_clip': 2.9635512828826904, 'timing/data': 0.0001947570126503706, 'timing/model': 1.2191296690143645, 'learning_rate/action_model': 9.296607284513752e-05, 'learning_rate/action_query_proj': 9.296607284513752e-05, 'learning_rate/base': 2.324151821128438e-05, 'epoch': 0.15}) +[2026-06-29 17:59:19,683][starVLA.training.train_starvla][INFO] - Step 350, Loss: {'train/loss': 0.818619966506958, 'train/grad_norm_pre_clip': 9.235239028930664, 'timing/data': 0.0002597800921648741, 'timing/model': 1.2181669299025089, 'learning_rate/action_model': 9.292672788899643e-05, 'learning_rate/action_query_proj': 9.292672788899643e-05, 'learning_rate/base': 2.3231681972249107e-05, 'epoch': 0.15}) +[2026-06-29 17:59:51,993][starVLA.training.train_starvla][INFO] - Step 351, Loss: {'train/loss': 0.056252621114254, 'train/grad_norm_pre_clip': 5.992969989776611, 'timing/data': 0.0001920920331031084, 'timing/model': 1.2208937800023705, 'learning_rate/action_model': 9.288728195022268e-05, 'learning_rate/action_query_proj': 9.288728195022268e-05, 'learning_rate/base': 2.322182048755567e-05, 'epoch': 0.15}) +[2026-06-29 18:00:24,345][starVLA.training.train_starvla][INFO] - Step 352, Loss: {'train/loss': 0.22799396514892578, 'train/grad_norm_pre_clip': 4.911896228790283, 'timing/data': 0.0003146929666399956, 'timing/model': 1.2218954958952963, 'learning_rate/action_model': 9.284773512614522e-05, 'learning_rate/action_query_proj': 9.284773512614522e-05, 'learning_rate/base': 2.3211933781536305e-05, 'epoch': 0.15}) +[2026-06-29 18:00:56,728][starVLA.training.train_starvla][INFO] - Step 353, Loss: {'train/loss': 0.06849144399166107, 'train/grad_norm_pre_clip': 3.388079881668091, 'timing/data': 0.00018991786055266857, 'timing/model': 1.2284870578441769, 'learning_rate/action_model': 9.280808751434187e-05, 'learning_rate/action_query_proj': 9.280808751434187e-05, 'learning_rate/base': 2.3202021878585467e-05, 'epoch': 0.15}) +[2026-06-29 18:01:29,093][starVLA.training.train_starvla][INFO] - Step 354, Loss: {'train/loss': 0.01361345686018467, 'train/grad_norm_pre_clip': 5.458770751953125, 'timing/data': 0.0002503320574760437, 'timing/model': 1.2260060322005302, 'learning_rate/action_model': 9.27683392126392e-05, 'learning_rate/action_query_proj': 9.27683392126392e-05, 'learning_rate/base': 2.31920848031598e-05, 'epoch': 0.15}) +[2026-06-29 18:02:01,408][starVLA.training.train_starvla][INFO] - Step 355, Loss: {'train/loss': 0.12753024697303772, 'train/grad_norm_pre_clip': 1.4707731008529663, 'timing/data': 0.00024882005527615547, 'timing/model': 1.221638707909733, 'learning_rate/action_model': 9.272849031911219e-05, 'learning_rate/action_query_proj': 9.272849031911219e-05, 'learning_rate/base': 2.3182122579778047e-05, 'epoch': 0.15}) +[2026-06-29 18:02:33,716][starVLA.training.train_starvla][INFO] - Step 356, Loss: {'train/loss': 0.0004398198798298836, 'train/grad_norm_pre_clip': 2.234839916229248, 'timing/data': 0.0001906000543385744, 'timing/model': 1.2190290989819914, 'learning_rate/action_model': 9.2688540932084e-05, 'learning_rate/action_query_proj': 9.2688540932084e-05, 'learning_rate/base': 2.3172135233021e-05, 'epoch': 0.15}) +[2026-06-29 18:03:06,041][starVLA.training.train_starvla][INFO] - Step 357, Loss: {'train/loss': 0.2532121241092682, 'train/grad_norm_pre_clip': 3.7361502647399902, 'timing/data': 0.00029085902497172356, 'timing/model': 1.2215384980663657, 'learning_rate/action_model': 9.26484911501258e-05, 'learning_rate/action_query_proj': 9.26484911501258e-05, 'learning_rate/base': 2.316212278753145e-05, 'epoch': 0.15}) +[2026-06-29 18:03:38,397][starVLA.training.train_starvla][INFO] - Step 358, Loss: {'train/loss': 0.041543856263160706, 'train/grad_norm_pre_clip': 4.1013875007629395, 'timing/data': 0.00019346480257809162, 'timing/model': 1.2228301488794386, 'learning_rate/action_model': 9.260834107205643e-05, 'learning_rate/action_query_proj': 9.260834107205643e-05, 'learning_rate/base': 2.3152085268014107e-05, 'epoch': 0.15}) +[2026-06-29 18:04:10,740][starVLA.training.train_starvla][INFO] - Step 359, Loss: {'train/loss': 0.0006452349480241537, 'train/grad_norm_pre_clip': 5.273508548736572, 'timing/data': 0.000195337925106287, 'timing/model': 1.217619040980935, 'learning_rate/action_model': 9.256809079694222e-05, 'learning_rate/action_query_proj': 9.256809079694222e-05, 'learning_rate/base': 2.3142022699235554e-05, 'epoch': 0.15}) +[2026-06-29 18:04:43,078][starVLA.training.train_starvla][INFO] - Step 360, Loss: {'train/loss': 0.12254086136817932, 'train/grad_norm_pre_clip': 2.261455774307251, 'timing/data': 0.00030668918043375015, 'timing/model': 1.219669893849641, 'learning_rate/action_model': 9.252774042409673e-05, 'learning_rate/action_query_proj': 9.252774042409673e-05, 'learning_rate/base': 2.313193510602418e-05, 'epoch': 0.15}) +[2026-06-29 18:05:15,462][starVLA.training.train_starvla][INFO] - Step 361, Loss: {'train/loss': 0.003708894830197096, 'train/grad_norm_pre_clip': 6.272435188293457, 'timing/data': 0.0002366460394114256, 'timing/model': 1.2187672350555658, 'learning_rate/action_model': 9.248729005308048e-05, 'learning_rate/action_query_proj': 9.248729005308048e-05, 'learning_rate/base': 2.312182251327012e-05, 'epoch': 0.16}) +[2026-06-29 18:05:47,836][starVLA.training.train_starvla][INFO] - Step 362, Loss: {'train/loss': 4.440531483851373e-06, 'train/grad_norm_pre_clip': 2.0674450397491455, 'timing/data': 0.00019467808306217194, 'timing/model': 1.2237275058869272, 'learning_rate/action_model': 9.244673978370076e-05, 'learning_rate/action_query_proj': 9.244673978370076e-05, 'learning_rate/base': 2.311168494592519e-05, 'epoch': 0.16}) +[2026-06-29 18:06:20,187][starVLA.training.train_starvla][INFO] - Step 363, Loss: {'train/loss': 0.0010872067650780082, 'train/grad_norm_pre_clip': 3.8655707836151123, 'timing/data': 0.00019421684555709362, 'timing/model': 1.220160388853401, 'learning_rate/action_model': 9.24060897160113e-05, 'learning_rate/action_query_proj': 9.24060897160113e-05, 'learning_rate/base': 2.3101522429002824e-05, 'epoch': 0.16}) +[2026-06-29 18:06:52,513][starVLA.training.train_starvla][INFO] - Step 364, Loss: {'train/loss': 0.021266978234052658, 'train/grad_norm_pre_clip': 2.5564186573028564, 'timing/data': 0.0002084430307149887, 'timing/model': 1.2210195418447256, 'learning_rate/action_model': 9.236533995031212e-05, 'learning_rate/action_query_proj': 9.236533995031212e-05, 'learning_rate/base': 2.309133498757803e-05, 'epoch': 0.16}) +[2026-06-29 18:07:24,842][starVLA.training.train_starvla][INFO] - Step 365, Loss: {'train/loss': 0.36871784925460815, 'train/grad_norm_pre_clip': 4.523277282714844, 'timing/data': 0.00024768710136413574, 'timing/model': 1.2212760590482503, 'learning_rate/action_model': 9.232449058714921e-05, 'learning_rate/action_query_proj': 9.232449058714921e-05, 'learning_rate/base': 2.3081122646787303e-05, 'epoch': 0.16}) +[2026-06-29 18:07:57,192][starVLA.training.train_starvla][INFO] - Step 366, Loss: {'train/loss': 0.01624230667948723, 'train/grad_norm_pre_clip': 6.610591411590576, 'timing/data': 0.00019291508942842484, 'timing/model': 1.219903222983703, 'learning_rate/action_model': 9.228354172731432e-05, 'learning_rate/action_query_proj': 9.228354172731432e-05, 'learning_rate/base': 2.307088543182858e-05, 'epoch': 0.16}) +[2026-06-29 18:08:29,671][starVLA.training.train_starvla][INFO] - Step 367, Loss: {'train/loss': 2.306613532709889e-05, 'train/grad_norm_pre_clip': 4.459617614746094, 'timing/data': 0.00019203289411962032, 'timing/model': 1.2246950957924128, 'learning_rate/action_model': 9.224249347184469e-05, 'learning_rate/action_query_proj': 9.224249347184469e-05, 'learning_rate/base': 2.3060623367961172e-05, 'epoch': 0.16}) +[2026-06-29 18:09:02,005][starVLA.training.train_starvla][INFO] - Step 368, Loss: {'train/loss': 0.21641644835472107, 'train/grad_norm_pre_clip': 5.999032497406006, 'timing/data': 0.00019858591258525848, 'timing/model': 1.2175581068731844, 'learning_rate/action_model': 9.220134592202282e-05, 'learning_rate/action_query_proj': 9.220134592202282e-05, 'learning_rate/base': 2.3050336480505704e-05, 'epoch': 0.16}) +[2026-06-29 18:09:34,309][starVLA.training.train_starvla][INFO] - Step 369, Loss: {'train/loss': 0.009428191930055618, 'train/grad_norm_pre_clip': 7.269495964050293, 'timing/data': 0.00019482895731925964, 'timing/model': 1.2226822071243078, 'learning_rate/action_model': 9.216009917937615e-05, 'learning_rate/action_query_proj': 9.216009917937615e-05, 'learning_rate/base': 2.3040024794844037e-05, 'epoch': 0.16}) +[2026-06-29 18:10:06,610][starVLA.training.train_starvla][INFO] - Step 370, Loss: {'train/loss': 0.08266693353652954, 'train/grad_norm_pre_clip': 4.110040187835693, 'timing/data': 0.00019930698908865452, 'timing/model': 1.2210634320508689, 'learning_rate/action_model': 9.211875334567698e-05, 'learning_rate/action_query_proj': 9.211875334567698e-05, 'learning_rate/base': 2.3029688336419245e-05, 'epoch': 0.16}) +[2026-06-29 18:10:38,949][starVLA.training.train_starvla][INFO] - Step 371, Loss: {'train/loss': 0.017862455919384956, 'train/grad_norm_pre_clip': 2.2430851459503174, 'timing/data': 0.00024694716557860374, 'timing/model': 1.2195839630439878, 'learning_rate/action_model': 9.207730852294201e-05, 'learning_rate/action_query_proj': 9.207730852294201e-05, 'learning_rate/base': 2.3019327130735502e-05, 'epoch': 0.16}) +[2026-06-29 18:11:11,246][starVLA.training.train_starvla][INFO] - Step 372, Loss: {'train/loss': 0.20084187388420105, 'train/grad_norm_pre_clip': 2.7938780784606934, 'timing/data': 0.00018730410374701023, 'timing/model': 1.2158543469849974, 'learning_rate/action_model': 9.203576481343223e-05, 'learning_rate/action_query_proj': 9.203576481343223e-05, 'learning_rate/base': 2.3008941203358057e-05, 'epoch': 0.16}) +[2026-06-29 18:11:43,566][starVLA.training.train_starvla][INFO] - Step 373, Loss: {'train/loss': 0.16406306624412537, 'train/grad_norm_pre_clip': 3.834176778793335, 'timing/data': 0.00019361590966582298, 'timing/model': 1.2176215378567576, 'learning_rate/action_model': 9.19941223196526e-05, 'learning_rate/action_query_proj': 9.19941223196526e-05, 'learning_rate/base': 2.299853057991315e-05, 'epoch': 0.16}) +[2026-06-29 18:12:15,839][starVLA.training.train_starvla][INFO] - Step 374, Loss: {'train/loss': 0.5826719403266907, 'train/grad_norm_pre_clip': 2.8551151752471924, 'timing/data': 0.00019259401597082615, 'timing/model': 1.223147044191137, 'learning_rate/action_model': 9.195238114435186e-05, 'learning_rate/action_query_proj': 9.195238114435186e-05, 'learning_rate/base': 2.2988095286087964e-05, 'epoch': 0.16}) +[2026-06-29 18:12:48,204][starVLA.training.train_starvla][INFO] - Step 375, Loss: {'train/loss': 0.04597729444503784, 'train/grad_norm_pre_clip': 2.5044641494750977, 'timing/data': 0.0002573269885033369, 'timing/model': 1.2208425882272422, 'learning_rate/action_model': 9.191054139052218e-05, 'learning_rate/action_query_proj': 9.191054139052218e-05, 'learning_rate/base': 2.2977635347630545e-05, 'epoch': 0.16}) +[2026-06-29 18:13:20,534][starVLA.training.train_starvla][INFO] - Step 376, Loss: {'train/loss': 0.0020252258982509375, 'train/grad_norm_pre_clip': 3.7689874172210693, 'timing/data': 0.00023421295918524265, 'timing/model': 1.2219929720740765, 'learning_rate/action_model': 9.186860316139901e-05, 'learning_rate/action_query_proj': 9.186860316139901e-05, 'learning_rate/base': 2.2967150790349752e-05, 'epoch': 0.16}) +[2026-06-29 18:13:52,864][starVLA.training.train_starvla][INFO] - Step 377, Loss: {'train/loss': 0.1115097627043724, 'train/grad_norm_pre_clip': 3.676309585571289, 'timing/data': 0.000260351924225688, 'timing/model': 1.2207654051017016, 'learning_rate/action_model': 9.182656656046078e-05, 'learning_rate/action_query_proj': 9.182656656046078e-05, 'learning_rate/base': 2.2956641640115194e-05, 'epoch': 0.16}) +[2026-06-29 18:14:25,209][starVLA.training.train_starvla][INFO] - Step 378, Loss: {'train/loss': 0.1294533610343933, 'train/grad_norm_pre_clip': 6.155693054199219, 'timing/data': 0.0002636078279465437, 'timing/model': 1.2224654438905418, 'learning_rate/action_model': 9.178443169142858e-05, 'learning_rate/action_query_proj': 9.178443169142858e-05, 'learning_rate/base': 2.2946107922857144e-05, 'epoch': 0.16}) +[2026-06-29 18:14:57,595][starVLA.training.train_starvla][INFO] - Step 379, Loss: {'train/loss': 0.09115638583898544, 'train/grad_norm_pre_clip': 1.6704859733581543, 'timing/data': 0.00025217700749635696, 'timing/model': 1.2190732329618186, 'learning_rate/action_model': 9.174219865826604e-05, 'learning_rate/action_query_proj': 9.174219865826604e-05, 'learning_rate/base': 2.293554966456651e-05, 'epoch': 0.16}) +[2026-06-29 18:15:29,963][starVLA.training.train_starvla][INFO] - Step 380, Loss: {'train/loss': 0.010636864230036736, 'train/grad_norm_pre_clip': 2.60659122467041, 'timing/data': 0.0002910001203417778, 'timing/model': 1.2217079049441963, 'learning_rate/action_model': 9.169986756517896e-05, 'learning_rate/action_query_proj': 9.169986756517896e-05, 'learning_rate/base': 2.292496689129474e-05, 'epoch': 0.16}) +[2026-06-29 18:16:02,351][starVLA.training.train_starvla][INFO] - Step 381, Loss: {'train/loss': 0.33567574620246887, 'train/grad_norm_pre_clip': 1.3283462524414062, 'timing/data': 0.00020200200378894806, 'timing/model': 1.2214824620168656, 'learning_rate/action_model': 9.165743851661513e-05, 'learning_rate/action_query_proj': 9.165743851661513e-05, 'learning_rate/base': 2.2914359629153783e-05, 'epoch': 0.16}) +[2026-06-29 18:16:34,726][starVLA.training.train_starvla][INFO] - Step 382, Loss: {'train/loss': 0.10109753161668777, 'train/grad_norm_pre_clip': 1.5318443775177002, 'timing/data': 0.0001984450500458479, 'timing/model': 1.2204448559787124, 'learning_rate/action_model': 9.161491161726398e-05, 'learning_rate/action_query_proj': 9.161491161726398e-05, 'learning_rate/base': 2.2903727904315996e-05, 'epoch': 0.16}) +[2026-06-29 18:17:07,065][starVLA.training.train_starvla][INFO] - Step 383, Loss: {'train/loss': 0.036138467490673065, 'train/grad_norm_pre_clip': 1.2365562915802002, 'timing/data': 0.0002442111726850271, 'timing/model': 1.2183373691514134, 'learning_rate/action_model': 9.157228697205643e-05, 'learning_rate/action_query_proj': 9.157228697205643e-05, 'learning_rate/base': 2.2893071743014106e-05, 'epoch': 0.16}) +[2026-06-29 18:17:39,425][starVLA.training.train_starvla][INFO] - Step 384, Loss: {'train/loss': 0.03769475221633911, 'train/grad_norm_pre_clip': 2.4761600494384766, 'timing/data': 0.00024517299607396126, 'timing/model': 1.2190117570571601, 'learning_rate/action_model': 9.152956468616451e-05, 'learning_rate/action_query_proj': 9.152956468616451e-05, 'learning_rate/base': 2.2882391171541128e-05, 'epoch': 0.17}) +[2026-06-29 18:18:11,793][starVLA.training.train_starvla][INFO] - Step 385, Loss: {'train/loss': 0.0018754785414785147, 'train/grad_norm_pre_clip': 2.083847999572754, 'timing/data': 0.0002525569871068001, 'timing/model': 1.2195745140779763, 'learning_rate/action_model': 9.148674486500127e-05, 'learning_rate/action_query_proj': 9.148674486500127e-05, 'learning_rate/base': 2.287168621625032e-05, 'epoch': 0.17}) +[2026-06-29 18:18:44,130][starVLA.training.train_starvla][INFO] - Step 386, Loss: {'train/loss': 0.3178727924823761, 'train/grad_norm_pre_clip': 2.704868793487549, 'timing/data': 0.00019423593766987324, 'timing/model': 1.2191296049859375, 'learning_rate/action_model': 9.144382761422033e-05, 'learning_rate/action_query_proj': 9.144382761422033e-05, 'learning_rate/base': 2.2860956903555083e-05, 'epoch': 0.17}) +[2026-06-29 18:19:16,491][starVLA.training.train_starvla][INFO] - Step 387, Loss: {'train/loss': 2.7716089334717253e-06, 'train/grad_norm_pre_clip': 1.2028093338012695, 'timing/data': 0.00023069605231285095, 'timing/model': 1.2193705101963133, 'learning_rate/action_model': 9.140081303971575e-05, 'learning_rate/action_query_proj': 9.140081303971575e-05, 'learning_rate/base': 2.2850203259928938e-05, 'epoch': 0.17}) +[2026-06-29 18:19:48,840][starVLA.training.train_starvla][INFO] - Step 388, Loss: {'train/loss': 0.018266554921865463, 'train/grad_norm_pre_clip': 5.720449447631836, 'timing/data': 0.00019088992848992348, 'timing/model': 1.2189903890248388, 'learning_rate/action_model': 9.135770124762172e-05, 'learning_rate/action_query_proj': 9.135770124762172e-05, 'learning_rate/base': 2.283942531190543e-05, 'epoch': 0.17}) +[2026-06-29 18:20:21,207][starVLA.training.train_starvla][INFO] - Step 389, Loss: {'train/loss': 6.829782068962231e-05, 'train/grad_norm_pre_clip': 4.500929832458496, 'timing/data': 0.00019142101518809795, 'timing/model': 1.2195706989150494, 'learning_rate/action_model': 9.131449234431229e-05, 'learning_rate/action_query_proj': 9.131449234431229e-05, 'learning_rate/base': 2.2828623086078073e-05, 'epoch': 0.17}) +[2026-06-29 18:20:53,569][starVLA.training.train_starvla][INFO] - Step 390, Loss: {'train/loss': 1.3411035979515873e-06, 'train/grad_norm_pre_clip': 2.4417808055877686, 'timing/data': 0.00019570998847484589, 'timing/model': 1.2209309190511703, 'learning_rate/action_model': 9.127118643640112e-05, 'learning_rate/action_query_proj': 9.127118643640112e-05, 'learning_rate/base': 2.281779660910028e-05, 'epoch': 0.17}) +[2026-06-29 18:21:25,915][starVLA.training.train_starvla][INFO] - Step 391, Loss: {'train/loss': 0.36317071318626404, 'train/grad_norm_pre_clip': 3.650557041168213, 'timing/data': 0.00019603990949690342, 'timing/model': 1.2185698358807713, 'learning_rate/action_model': 9.122778363074128e-05, 'learning_rate/action_query_proj': 9.122778363074128e-05, 'learning_rate/base': 2.280694590768532e-05, 'epoch': 0.17}) +[2026-06-29 18:21:58,241][starVLA.training.train_starvla][INFO] - Step 392, Loss: {'train/loss': 0.776508629322052, 'train/grad_norm_pre_clip': 5.862035274505615, 'timing/data': 0.00019619986414909363, 'timing/model': 1.220041446853429, 'learning_rate/action_model': 9.118428403442483e-05, 'learning_rate/action_query_proj': 9.118428403442483e-05, 'learning_rate/base': 2.2796071008606208e-05, 'epoch': 0.17}) +[2026-06-29 18:22:30,596][starVLA.training.train_starvla][INFO] - Step 393, Loss: {'train/loss': 0.004272446967661381, 'train/grad_norm_pre_clip': 2.2458529472351074, 'timing/data': 0.00022915308363735676, 'timing/model': 1.2181791989132762, 'learning_rate/action_model': 9.114068775478273e-05, 'learning_rate/action_query_proj': 9.114068775478273e-05, 'learning_rate/base': 2.2785171938695682e-05, 'epoch': 0.17}) +[2026-06-29 18:23:02,927][starVLA.training.train_starvla][INFO] - Step 394, Loss: {'train/loss': 0.17636941373348236, 'train/grad_norm_pre_clip': 5.436709880828857, 'timing/data': 0.000333509873598814, 'timing/model': 1.225640189833939, 'learning_rate/action_model': 9.109699489938446e-05, 'learning_rate/action_query_proj': 9.109699489938446e-05, 'learning_rate/base': 2.2774248724846114e-05, 'epoch': 0.17}) +[2026-06-29 18:23:35,337][starVLA.training.train_starvla][INFO] - Step 395, Loss: {'train/loss': 0.00669162068516016, 'train/grad_norm_pre_clip': 4.945994853973389, 'timing/data': 0.00019313395023345947, 'timing/model': 1.220358954044059, 'learning_rate/action_model': 9.10532055760378e-05, 'learning_rate/action_query_proj': 9.10532055760378e-05, 'learning_rate/base': 2.276330139400945e-05, 'epoch': 0.17}) +[2026-06-29 18:24:07,711][starVLA.training.train_starvla][INFO] - Step 396, Loss: {'train/loss': 0.1277191936969757, 'train/grad_norm_pre_clip': 4.6507792472839355, 'timing/data': 0.00020093913190066814, 'timing/model': 1.218407201115042, 'learning_rate/action_model': 9.100931989278851e-05, 'learning_rate/action_query_proj': 9.100931989278851e-05, 'learning_rate/base': 2.2752329973197127e-05, 'epoch': 0.17}) +[2026-06-29 18:24:40,085][starVLA.training.train_starvla][INFO] - Step 397, Loss: {'train/loss': 0.005530776455998421, 'train/grad_norm_pre_clip': 3.0853054523468018, 'timing/data': 0.00019644084386527538, 'timing/model': 1.221781007014215, 'learning_rate/action_model': 9.096533795792021e-05, 'learning_rate/action_query_proj': 9.096533795792021e-05, 'learning_rate/base': 2.2741334489480054e-05, 'epoch': 0.17}) +[2026-06-29 18:25:12,490][starVLA.training.train_starvla][INFO] - Step 398, Loss: {'train/loss': 0.17357641458511353, 'train/grad_norm_pre_clip': 2.1097075939178467, 'timing/data': 0.00020236195996403694, 'timing/model': 1.224298289977014, 'learning_rate/action_model': 9.092125987995393e-05, 'learning_rate/action_query_proj': 9.092125987995393e-05, 'learning_rate/base': 2.2730314969988483e-05, 'epoch': 0.17}) +[2026-06-29 18:25:44,841][starVLA.training.train_starvla][INFO] - Step 399, Loss: {'train/loss': 0.4257305860519409, 'train/grad_norm_pre_clip': 3.0709147453308105, 'timing/data': 0.0002902678679674864, 'timing/model': 1.2178881098516285, 'learning_rate/action_model': 9.087708576764794e-05, 'learning_rate/action_query_proj': 9.087708576764794e-05, 'learning_rate/base': 2.2719271441911985e-05, 'epoch': 0.17}) +[2026-06-29 18:26:17,191][starVLA.training.train_starvla][INFO] - Step 400, Loss: {'train/loss': 0.1165473610162735, 'train/grad_norm_pre_clip': 4.0115461349487305, 'timing/data': 0.00025322800502181053, 'timing/model': 1.2204690808430314, 'learning_rate/action_model': 9.083281572999748e-05, 'learning_rate/action_query_proj': 9.083281572999748e-05, 'learning_rate/base': 2.270820393249937e-05, 'epoch': 0.17}) +[2026-06-29 18:26:49,561][starVLA.training.train_starvla][INFO] - Step 401, Loss: {'train/loss': 0.32376593351364136, 'train/grad_norm_pre_clip': 2.562405824661255, 'timing/data': 0.00027887593023478985, 'timing/model': 1.221918567083776, 'learning_rate/action_model': 9.078844987623446e-05, 'learning_rate/action_query_proj': 9.078844987623446e-05, 'learning_rate/base': 2.2697112469058616e-05, 'epoch': 0.17}) +[2026-06-29 18:27:21,945][starVLA.training.train_starvla][INFO] - Step 402, Loss: {'train/loss': 0.015272648073732853, 'train/grad_norm_pre_clip': 1.9579050540924072, 'timing/data': 0.0002867721486836672, 'timing/model': 1.2227168278768659, 'learning_rate/action_model': 9.074398831582722e-05, 'learning_rate/action_query_proj': 9.074398831582722e-05, 'learning_rate/base': 2.2685997078956804e-05, 'epoch': 0.17}) +[2026-06-29 18:27:54,279][starVLA.training.train_starvla][INFO] - Step 403, Loss: {'train/loss': 0.4532875418663025, 'train/grad_norm_pre_clip': 8.737520217895508, 'timing/data': 0.00020066904835402966, 'timing/model': 1.2182358540594578, 'learning_rate/action_model': 9.069943115848024e-05, 'learning_rate/action_query_proj': 9.069943115848024e-05, 'learning_rate/base': 2.267485778962006e-05, 'epoch': 0.17}) +[2026-06-29 18:28:26,628][starVLA.training.train_starvla][INFO] - Step 404, Loss: {'train/loss': 0.008980650454759598, 'train/grad_norm_pre_clip': 3.8993871212005615, 'timing/data': 0.00019489810802042484, 'timing/model': 1.216446615057066, 'learning_rate/action_model': 9.065477851413386e-05, 'learning_rate/action_query_proj': 9.065477851413386e-05, 'learning_rate/base': 2.2663694628533465e-05, 'epoch': 0.17}) +[2026-06-29 18:28:59,094][starVLA.training.train_starvla][INFO] - Step 405, Loss: {'train/loss': 0.026419715955853462, 'train/grad_norm_pre_clip': 2.794712781906128, 'timing/data': 0.00025473092682659626, 'timing/model': 1.2232891959138215, 'learning_rate/action_model': 9.061003049296406e-05, 'learning_rate/action_query_proj': 9.061003049296406e-05, 'learning_rate/base': 2.2652507623241015e-05, 'epoch': 0.17}) +[2026-06-29 18:29:31,446][starVLA.training.train_starvla][INFO] - Step 406, Loss: {'train/loss': 0.11927111446857452, 'train/grad_norm_pre_clip': 1.682958722114563, 'timing/data': 0.0001935351174324751, 'timing/model': 1.2215399558190256, 'learning_rate/action_model': 9.056518720538213e-05, 'learning_rate/action_query_proj': 9.056518720538213e-05, 'learning_rate/base': 2.2641296801345534e-05, 'epoch': 0.17}) +[2026-06-29 18:30:03,770][starVLA.training.train_starvla][INFO] - Step 407, Loss: {'train/loss': 0.035832326859235764, 'train/grad_norm_pre_clip': 2.691004753112793, 'timing/data': 0.0003848560154438019, 'timing/model': 1.2188827251084149, 'learning_rate/action_model': 9.052024876203442e-05, 'learning_rate/action_query_proj': 9.052024876203442e-05, 'learning_rate/base': 2.2630062190508604e-05, 'epoch': 0.18}) +[2026-06-29 18:30:36,137][starVLA.training.train_starvla][INFO] - Step 408, Loss: {'train/loss': 0.11524161696434021, 'train/grad_norm_pre_clip': 3.3778603076934814, 'timing/data': 0.0001903190277516842, 'timing/model': 1.2187315779738128, 'learning_rate/action_model': 9.047521527380208e-05, 'learning_rate/action_query_proj': 9.047521527380208e-05, 'learning_rate/base': 2.261880381845052e-05, 'epoch': 0.18}) +[2026-06-29 18:31:08,480][starVLA.training.train_starvla][INFO] - Step 409, Loss: {'train/loss': 0.04880670830607414, 'train/grad_norm_pre_clip': 1.842861533164978, 'timing/data': 0.0002440209500491619, 'timing/model': 1.222417065873742, 'learning_rate/action_model': 9.043008685180075e-05, 'learning_rate/action_query_proj': 9.043008685180075e-05, 'learning_rate/base': 2.2607521712950187e-05, 'epoch': 0.18}) +[2026-06-29 18:31:40,849][starVLA.training.train_starvla][INFO] - Step 410, Loss: {'train/loss': 0.13544835150241852, 'train/grad_norm_pre_clip': 2.1971635818481445, 'timing/data': 0.00024405098520219326, 'timing/model': 1.2207376980222762, 'learning_rate/action_model': 9.038486360738036e-05, 'learning_rate/action_query_proj': 9.038486360738036e-05, 'learning_rate/base': 2.259621590184509e-05, 'epoch': 0.18}) +[2026-06-29 18:32:13,215][starVLA.training.train_starvla][INFO] - Step 411, Loss: {'train/loss': 0.0007421316113322973, 'train/grad_norm_pre_clip': 3.0237081050872803, 'timing/data': 0.0002513348590582609, 'timing/model': 1.2203970479313284, 'learning_rate/action_model': 9.033954565212471e-05, 'learning_rate/action_query_proj': 9.033954565212471e-05, 'learning_rate/base': 2.2584886413031178e-05, 'epoch': 0.18}) +[2026-06-29 18:32:45,585][starVLA.training.train_starvla][INFO] - Step 412, Loss: {'train/loss': 0.21543490886688232, 'train/grad_norm_pre_clip': 4.36279296875, 'timing/data': 0.000251776073127985, 'timing/model': 1.2223901259712875, 'learning_rate/action_model': 9.029413309785141e-05, 'learning_rate/action_query_proj': 9.029413309785141e-05, 'learning_rate/base': 2.2573533274462853e-05, 'epoch': 0.18}) +[2026-06-29 18:33:17,931][starVLA.training.train_starvla][INFO] - Step 413, Loss: {'train/loss': 0.0182648878544569, 'train/grad_norm_pre_clip': 1.643050193786621, 'timing/data': 0.00020728097297251225, 'timing/model': 1.2204198229592294, 'learning_rate/action_model': 9.02486260566114e-05, 'learning_rate/action_query_proj': 9.02486260566114e-05, 'learning_rate/base': 2.256215651415285e-05, 'epoch': 0.18}) +[2026-06-29 18:33:50,268][starVLA.training.train_starvla][INFO] - Step 414, Loss: {'train/loss': 0.06739908456802368, 'train/grad_norm_pre_clip': 1.3253921270370483, 'timing/data': 0.00021797115914523602, 'timing/model': 1.2241589999757707, 'learning_rate/action_model': 9.020302464068876e-05, 'learning_rate/action_query_proj': 9.020302464068876e-05, 'learning_rate/base': 2.255075616017219e-05, 'epoch': 0.18}) +[2026-06-29 18:34:22,653][starVLA.training.train_starvla][INFO] - Step 415, Loss: {'train/loss': 0.00555684557184577, 'train/grad_norm_pre_clip': 1.393851399421692, 'timing/data': 0.00020141992717981339, 'timing/model': 1.2226196029223502, 'learning_rate/action_model': 9.015732896260048e-05, 'learning_rate/action_query_proj': 9.015732896260048e-05, 'learning_rate/base': 2.253933224065012e-05, 'epoch': 0.18}) +[2026-06-29 18:34:55,002][starVLA.training.train_starvla][INFO] - Step 416, Loss: {'train/loss': 0.07477252930402756, 'train/grad_norm_pre_clip': 1.4369450807571411, 'timing/data': 0.00034045311622321606, 'timing/model': 1.2236159548629075, 'learning_rate/action_model': 9.01115391350961e-05, 'learning_rate/action_query_proj': 9.01115391350961e-05, 'learning_rate/base': 2.2527884783774024e-05, 'epoch': 0.18}) +[2026-06-29 18:35:27,364][starVLA.training.train_starvla][INFO] - Step 417, Loss: {'train/loss': 0.021014781668782234, 'train/grad_norm_pre_clip': 1.8558733463287354, 'timing/data': 0.00020113005302846432, 'timing/model': 1.221554656047374, 'learning_rate/action_model': 9.006565527115746e-05, 'learning_rate/action_query_proj': 9.006565527115746e-05, 'learning_rate/base': 2.2516413817789364e-05, 'epoch': 0.18}) +[2026-06-29 18:35:59,697][starVLA.training.train_starvla][INFO] - Step 418, Loss: {'train/loss': 0.00041513892938382924, 'train/grad_norm_pre_clip': 1.7301208972930908, 'timing/data': 0.0002212170511484146, 'timing/model': 1.219500461127609, 'learning_rate/action_model': 9.001967748399842e-05, 'learning_rate/action_query_proj': 9.001967748399842e-05, 'learning_rate/base': 2.2504919370999605e-05, 'epoch': 0.18}) +[2026-06-29 18:36:32,040][starVLA.training.train_starvla][INFO] - Step 419, Loss: {'train/loss': 0.003921802155673504, 'train/grad_norm_pre_clip': 2.3195176124572754, 'timing/data': 0.00020064786076545715, 'timing/model': 1.2199660539627075, 'learning_rate/action_model': 8.997360588706464e-05, 'learning_rate/action_query_proj': 8.997360588706464e-05, 'learning_rate/base': 2.249340147176616e-05, 'epoch': 0.18}) +[2026-06-29 18:37:04,402][starVLA.training.train_starvla][INFO] - Step 420, Loss: {'train/loss': 0.2892816960811615, 'train/grad_norm_pre_clip': 3.496497392654419, 'timing/data': 0.0002901081461459398, 'timing/model': 1.220244905911386, 'learning_rate/action_model': 8.992744059403314e-05, 'learning_rate/action_query_proj': 8.992744059403314e-05, 'learning_rate/base': 2.2481860148508284e-05, 'epoch': 0.18}) +[2026-06-29 18:37:36,739][starVLA.training.train_starvla][INFO] - Step 421, Loss: {'train/loss': 0.0030945672187954187, 'train/grad_norm_pre_clip': 3.6398026943206787, 'timing/data': 0.00028356490656733513, 'timing/model': 1.2194678538944572, 'learning_rate/action_model': 8.988118171881225e-05, 'learning_rate/action_query_proj': 8.988118171881225e-05, 'learning_rate/base': 2.2470295429703064e-05, 'epoch': 0.18}) +[2026-06-29 18:38:09,070][starVLA.training.train_starvla][INFO] - Step 422, Loss: {'train/loss': 0.12493229657411575, 'train/grad_norm_pre_clip': 1.3877806663513184, 'timing/data': 0.0002857090439647436, 'timing/model': 1.219925896031782, 'learning_rate/action_model': 8.983482937554113e-05, 'learning_rate/action_query_proj': 8.983482937554113e-05, 'learning_rate/base': 2.245870734388528e-05, 'epoch': 0.18}) +[2026-06-29 18:38:41,410][starVLA.training.train_starvla][INFO] - Step 423, Loss: {'train/loss': 0.10132284462451935, 'train/grad_norm_pre_clip': 4.689676761627197, 'timing/data': 0.0003378379624336958, 'timing/model': 1.2211262800265104, 'learning_rate/action_model': 8.978838367858957e-05, 'learning_rate/action_query_proj': 8.978838367858957e-05, 'learning_rate/base': 2.244709591964739e-05, 'epoch': 0.18}) +[2026-06-29 18:39:13,724][starVLA.training.train_starvla][INFO] - Step 424, Loss: {'train/loss': 0.14519944787025452, 'train/grad_norm_pre_clip': 4.377133369445801, 'timing/data': 0.00032785884104669094, 'timing/model': 1.2200583219528198, 'learning_rate/action_model': 8.974184474255772e-05, 'learning_rate/action_query_proj': 8.974184474255772e-05, 'learning_rate/base': 2.243546118563943e-05, 'epoch': 0.18}) +[2026-06-29 18:39:46,089][starVLA.training.train_starvla][INFO] - Step 425, Loss: {'train/loss': 0.0033244637306779623, 'train/grad_norm_pre_clip': 6.541028022766113, 'timing/data': 0.0001934750471264124, 'timing/model': 1.2206028189975768, 'learning_rate/action_model': 8.969521268227575e-05, 'learning_rate/action_query_proj': 8.969521268227575e-05, 'learning_rate/base': 2.2423803170568937e-05, 'epoch': 0.18}) +[2026-06-29 18:40:18,453][starVLA.training.train_starvla][INFO] - Step 426, Loss: {'train/loss': 0.00013926053361501545, 'train/grad_norm_pre_clip': 2.941195249557495, 'timing/data': 0.00019935681484639645, 'timing/model': 1.2235472158063203, 'learning_rate/action_model': 8.964848761280368e-05, 'learning_rate/action_query_proj': 8.964848761280368e-05, 'learning_rate/base': 2.241212190320092e-05, 'epoch': 0.18}) +[2026-06-29 18:40:50,792][starVLA.training.train_starvla][INFO] - Step 427, Loss: {'train/loss': 0.0016076939646154642, 'train/grad_norm_pre_clip': 3.9065868854522705, 'timing/data': 0.0001972629688680172, 'timing/model': 1.2219251850619912, 'learning_rate/action_model': 8.960166964943096e-05, 'learning_rate/action_query_proj': 8.960166964943096e-05, 'learning_rate/base': 2.240041741235774e-05, 'epoch': 0.18}) +[2026-06-29 18:41:23,140][starVLA.training.train_starvla][INFO] - Step 428, Loss: {'train/loss': 0.15065760910511017, 'train/grad_norm_pre_clip': 4.55888032913208, 'timing/data': 0.00019901595078408718, 'timing/model': 1.2204732662066817, 'learning_rate/action_model': 8.955475890767623e-05, 'learning_rate/action_query_proj': 8.955475890767623e-05, 'learning_rate/base': 2.238868972691906e-05, 'epoch': 0.18}) +[2026-06-29 18:41:55,527][starVLA.training.train_starvla][INFO] - Step 429, Loss: {'train/loss': 0.5719553232192993, 'train/grad_norm_pre_clip': 4.270114421844482, 'timing/data': 0.00023395195603370667, 'timing/model': 1.2181710221339017, 'learning_rate/action_model': 8.950775550328711e-05, 'learning_rate/action_query_proj': 8.950775550328711e-05, 'learning_rate/base': 2.2376938875821777e-05, 'epoch': 0.18}) +[2026-06-29 18:42:27,887][starVLA.training.train_starvla][INFO] - Step 430, Loss: {'train/loss': 0.004593044053763151, 'train/grad_norm_pre_clip': 3.6766271591186523, 'timing/data': 0.0001995260827243328, 'timing/model': 1.2195851460564882, 'learning_rate/action_model': 8.946065955223984e-05, 'learning_rate/action_query_proj': 8.946065955223984e-05, 'learning_rate/base': 2.236516488805996e-05, 'epoch': 0.19}) +[2026-06-29 18:43:00,272][starVLA.training.train_starvla][INFO] - Step 431, Loss: {'train/loss': 0.0005244986969046295, 'train/grad_norm_pre_clip': 2.9827423095703125, 'timing/data': 0.00020691100507974625, 'timing/model': 1.2215249680448323, 'learning_rate/action_model': 8.941347117073896e-05, 'learning_rate/action_query_proj': 8.941347117073896e-05, 'learning_rate/base': 2.235336779268474e-05, 'epoch': 0.19}) +[2026-06-29 18:43:32,624][starVLA.training.train_starvla][INFO] - Step 432, Loss: {'train/loss': 0.021212786436080933, 'train/grad_norm_pre_clip': 2.184739351272583, 'timing/data': 0.00019867485389113426, 'timing/model': 1.2235453601460904, 'learning_rate/action_model': 8.936619047521714e-05, 'learning_rate/action_query_proj': 8.936619047521714e-05, 'learning_rate/base': 2.2341547618804285e-05, 'epoch': 0.19}) +[2026-06-29 18:44:04,924][starVLA.training.train_starvla][INFO] - Step 433, Loss: {'train/loss': 0.014664851129055023, 'train/grad_norm_pre_clip': 2.9823272228240967, 'timing/data': 0.00019718194380402565, 'timing/model': 1.2219568288419396, 'learning_rate/action_model': 8.931881758233477e-05, 'learning_rate/action_query_proj': 8.931881758233477e-05, 'learning_rate/base': 2.2329704395583694e-05, 'epoch': 0.19}) +[2026-06-29 18:44:37,271][starVLA.training.train_starvla][INFO] - Step 434, Loss: {'train/loss': 0.04260702803730965, 'train/grad_norm_pre_clip': 2.0347907543182373, 'timing/data': 0.00024869898334145546, 'timing/model': 1.22057089000009, 'learning_rate/action_model': 8.927135260897978e-05, 'learning_rate/action_query_proj': 8.927135260897978e-05, 'learning_rate/base': 2.2317838152244945e-05, 'epoch': 0.19}) +[2026-06-29 18:45:09,670][starVLA.training.train_starvla][INFO] - Step 435, Loss: {'train/loss': 0.00024599992320872843, 'train/grad_norm_pre_clip': 4.439006805419922, 'timing/data': 0.0003340609837323427, 'timing/model': 1.224397815996781, 'learning_rate/action_model': 8.922379567226726e-05, 'learning_rate/action_query_proj': 8.922379567226726e-05, 'learning_rate/base': 2.2305948918066815e-05, 'epoch': 0.19}) +[2026-06-29 18:45:42,019][starVLA.training.train_starvla][INFO] - Step 436, Loss: {'train/loss': 0.0028003999032080173, 'train/grad_norm_pre_clip': 1.436566710472107, 'timing/data': 0.00020404509268701077, 'timing/model': 1.2165628930088133, 'learning_rate/action_model': 8.917614688953923e-05, 'learning_rate/action_query_proj': 8.917614688953923e-05, 'learning_rate/base': 2.2294036722384807e-05, 'epoch': 0.19}) +[2026-06-29 18:46:14,304][starVLA.training.train_starvla][INFO] - Step 437, Loss: {'train/loss': 0.0055403816513717175, 'train/grad_norm_pre_clip': 1.8349108695983887, 'timing/data': 0.0003443099558353424, 'timing/model': 1.2195585980080068, 'learning_rate/action_model': 8.912840637836433e-05, 'learning_rate/action_query_proj': 8.912840637836433e-05, 'learning_rate/base': 2.2282101594591082e-05, 'epoch': 0.19}) +[2026-06-29 18:46:46,636][starVLA.training.train_starvla][INFO] - Step 438, Loss: {'train/loss': 0.6765822768211365, 'train/grad_norm_pre_clip': 2.4530653953552246, 'timing/data': 0.00019683199934661388, 'timing/model': 1.2190141209866852, 'learning_rate/action_model': 8.90805742565375e-05, 'learning_rate/action_query_proj': 8.90805742565375e-05, 'learning_rate/base': 2.2270143564134375e-05, 'epoch': 0.19}) +[2026-06-29 18:47:19,006][starVLA.training.train_starvla][INFO] - Step 439, Loss: {'train/loss': 0.00036765943514183164, 'train/grad_norm_pre_clip': 3.5284345149993896, 'timing/data': 0.00024915020912885666, 'timing/model': 1.2228709780611098, 'learning_rate/action_model': 8.903265064207976e-05, 'learning_rate/action_query_proj': 8.903265064207976e-05, 'learning_rate/base': 2.225816266051994e-05, 'epoch': 0.19}) +[2026-06-29 18:47:51,346][starVLA.training.train_starvla][INFO] - Step 440, Loss: {'train/loss': 0.07613802701234818, 'train/grad_norm_pre_clip': 3.899545431137085, 'timing/data': 0.00019605993293225765, 'timing/model': 1.2188994409516454, 'learning_rate/action_model': 8.898463565323789e-05, 'learning_rate/action_query_proj': 8.898463565323789e-05, 'learning_rate/base': 2.2246158913309473e-05, 'epoch': 0.19}) +[2026-06-29 18:48:23,682][starVLA.training.train_starvla][INFO] - Step 441, Loss: {'train/loss': 0.031291570514440536, 'train/grad_norm_pre_clip': 1.5816993713378906, 'timing/data': 0.00029478687793016434, 'timing/model': 1.2197027530055493, 'learning_rate/action_model': 8.893652940848407e-05, 'learning_rate/action_query_proj': 8.893652940848407e-05, 'learning_rate/base': 2.2234132352121017e-05, 'epoch': 0.19}) +[2026-06-29 18:48:56,027][starVLA.training.train_starvla][INFO] - Step 442, Loss: {'train/loss': 0.18585674464702606, 'train/grad_norm_pre_clip': 1.1887707710266113, 'timing/data': 0.00023309001699090004, 'timing/model': 1.2208886069711298, 'learning_rate/action_model': 8.888833202651569e-05, 'learning_rate/action_query_proj': 8.888833202651569e-05, 'learning_rate/base': 2.2222083006628923e-05, 'epoch': 0.19}) +[2026-06-29 18:49:28,395][starVLA.training.train_starvla][INFO] - Step 443, Loss: {'train/loss': 0.8458645939826965, 'train/grad_norm_pre_clip': 2.7255406379699707, 'timing/data': 0.00022349180653691292, 'timing/model': 1.2210743420291692, 'learning_rate/action_model': 8.884004362625498e-05, 'learning_rate/action_query_proj': 8.884004362625498e-05, 'learning_rate/base': 2.2210010906563744e-05, 'epoch': 0.19}) +[2026-06-29 18:50:00,911][starVLA.training.train_starvla][INFO] - Step 444, Loss: {'train/loss': 0.1527869701385498, 'train/grad_norm_pre_clip': 3.023834228515625, 'timing/data': 0.00019749300554394722, 'timing/model': 1.2218064879998565, 'learning_rate/action_model': 8.87916643268488e-05, 'learning_rate/action_query_proj': 8.87916643268488e-05, 'learning_rate/base': 2.21979160817122e-05, 'epoch': 0.19}) +[2026-06-29 18:50:33,274][starVLA.training.train_starvla][INFO] - Step 445, Loss: {'train/loss': 0.20515023171901703, 'train/grad_norm_pre_clip': 1.0963172912597656, 'timing/data': 0.0002017100341618061, 'timing/model': 1.21916500502266, 'learning_rate/action_model': 8.874319424766824e-05, 'learning_rate/action_query_proj': 8.874319424766824e-05, 'learning_rate/base': 2.218579856191706e-05, 'epoch': 0.19}) +[2026-06-29 18:51:05,621][starVLA.training.train_starvla][INFO] - Step 446, Loss: {'train/loss': 0.09489144384860992, 'train/grad_norm_pre_clip': 1.582398533821106, 'timing/data': 0.0001927840057760477, 'timing/model': 1.220600196160376, 'learning_rate/action_model': 8.86946335083084e-05, 'learning_rate/action_query_proj': 8.86946335083084e-05, 'learning_rate/base': 2.21736583770771e-05, 'epoch': 0.19}) +[2026-06-29 18:51:37,979][starVLA.training.train_starvla][INFO] - Step 447, Loss: {'train/loss': 0.3989012539386749, 'train/grad_norm_pre_clip': 4.352102279663086, 'timing/data': 0.00019219191744923592, 'timing/model': 1.220224387012422, 'learning_rate/action_model': 8.86459822285881e-05, 'learning_rate/action_query_proj': 8.86459822285881e-05, 'learning_rate/base': 2.2161495557147025e-05, 'epoch': 0.19}) +[2026-06-29 18:52:10,318][starVLA.training.train_starvla][INFO] - Step 448, Loss: {'train/loss': 0.14645445346832275, 'train/grad_norm_pre_clip': 2.251650094985962, 'timing/data': 0.0001878549810498953, 'timing/model': 1.2210884189698845, 'learning_rate/action_model': 8.85972405285495e-05, 'learning_rate/action_query_proj': 8.85972405285495e-05, 'learning_rate/base': 2.2149310132137376e-05, 'epoch': 0.19}) +[2026-06-29 18:52:42,675][starVLA.training.train_starvla][INFO] - Step 449, Loss: {'train/loss': 0.5036839246749878, 'train/grad_norm_pre_clip': 2.284111738204956, 'timing/data': 0.0002493311185389757, 'timing/model': 1.224033216945827, 'learning_rate/action_model': 8.854840852845794e-05, 'learning_rate/action_query_proj': 8.854840852845794e-05, 'learning_rate/base': 2.2137102132114484e-05, 'epoch': 0.19}) +[2026-06-29 18:53:15,041][starVLA.training.train_starvla][INFO] - Step 450, Loss: {'train/loss': 0.295350044965744, 'train/grad_norm_pre_clip': 2.4843013286590576, 'timing/data': 0.00018959795124828815, 'timing/model': 1.2284446919802576, 'learning_rate/action_model': 8.84994863488015e-05, 'learning_rate/action_query_proj': 8.84994863488015e-05, 'learning_rate/base': 2.2124871587200376e-05, 'epoch': 0.19}) +[2026-06-29 18:53:47,365][starVLA.training.train_starvla][INFO] - Step 451, Loss: {'train/loss': 0.5073750615119934, 'train/grad_norm_pre_clip': 2.198784112930298, 'timing/data': 0.00019188201986253262, 'timing/model': 1.2220981118734926, 'learning_rate/action_model': 8.845047411029078e-05, 'learning_rate/action_query_proj': 8.845047411029078e-05, 'learning_rate/base': 2.2112618527572694e-05, 'epoch': 0.19}) +[2026-06-29 18:54:19,732][starVLA.training.train_starvla][INFO] - Step 452, Loss: {'train/loss': 0.24510183930397034, 'train/grad_norm_pre_clip': 1.3435229063034058, 'timing/data': 0.00023334077559411526, 'timing/model': 1.218726116931066, 'learning_rate/action_model': 8.840137193385865e-05, 'learning_rate/action_query_proj': 8.840137193385865e-05, 'learning_rate/base': 2.2100342983464663e-05, 'epoch': 0.19}) +[2026-06-29 18:54:52,101][starVLA.training.train_starvla][INFO] - Step 453, Loss: {'train/loss': 0.005341336131095886, 'train/grad_norm_pre_clip': 2.07243275642395, 'timing/data': 0.00019152206368744373, 'timing/model': 1.2215101190377027, 'learning_rate/action_model': 8.835217994065982e-05, 'learning_rate/action_query_proj': 8.835217994065982e-05, 'learning_rate/base': 2.2088044985164956e-05, 'epoch': 0.19}) +[2026-06-29 18:55:24,483][starVLA.training.train_starvla][INFO] - Step 454, Loss: {'train/loss': 3.796740566031076e-05, 'train/grad_norm_pre_clip': 3.621497392654419, 'timing/data': 0.00024214689619839191, 'timing/model': 1.2216858081519604, 'learning_rate/action_model': 8.830289825207065e-05, 'learning_rate/action_query_proj': 8.830289825207065e-05, 'learning_rate/base': 2.2075724563017662e-05, 'epoch': 0.2}) +[2026-06-29 18:55:56,835][starVLA.training.train_starvla][INFO] - Step 455, Loss: {'train/loss': 0.45572835206985474, 'train/grad_norm_pre_clip': 6.394002914428711, 'timing/data': 0.00024832901544868946, 'timing/model': 1.2186429919674993, 'learning_rate/action_model': 8.825352698968882e-05, 'learning_rate/action_query_proj': 8.825352698968882e-05, 'learning_rate/base': 2.2063381747422206e-05, 'epoch': 0.2}) +[2026-06-29 18:56:29,220][starVLA.training.train_starvla][INFO] - Step 456, Loss: {'train/loss': 0.13306762278079987, 'train/grad_norm_pre_clip': 1.827221393585205, 'timing/data': 0.00018761400133371353, 'timing/model': 1.2205244190990925, 'learning_rate/action_model': 8.820406627533298e-05, 'learning_rate/action_query_proj': 8.820406627533298e-05, 'learning_rate/base': 2.2051016568833246e-05, 'epoch': 0.2}) +[2026-06-29 18:57:01,603][starVLA.training.train_starvla][INFO] - Step 457, Loss: {'train/loss': 0.0006209022831171751, 'train/grad_norm_pre_clip': 2.347115993499756, 'timing/data': 0.00038168998435139656, 'timing/model': 1.2213733370881528, 'learning_rate/action_model': 8.815451623104256e-05, 'learning_rate/action_query_proj': 8.815451623104256e-05, 'learning_rate/base': 2.203862905776064e-05, 'epoch': 0.2}) +[2026-06-29 18:57:33,963][starVLA.training.train_starvla][INFO] - Step 458, Loss: {'train/loss': 0.29338565468788147, 'train/grad_norm_pre_clip': 3.9087018966674805, 'timing/data': 0.0001884459052234888, 'timing/model': 1.2272936368826777, 'learning_rate/action_model': 8.810487697907735e-05, 'learning_rate/action_query_proj': 8.810487697907735e-05, 'learning_rate/base': 2.2026219244769338e-05, 'epoch': 0.2}) +[2026-06-29 18:58:06,320][starVLA.training.train_starvla][INFO] - Step 459, Loss: {'train/loss': 0.6030211448669434, 'train/grad_norm_pre_clip': 4.368103981018066, 'timing/data': 0.00032432214356958866, 'timing/model': 1.220269203884527, 'learning_rate/action_model': 8.805514864191728e-05, 'learning_rate/action_query_proj': 8.805514864191728e-05, 'learning_rate/base': 2.201378716047932e-05, 'epoch': 0.2}) +[2026-06-29 18:58:38,651][starVLA.training.train_starvla][INFO] - Step 460, Loss: {'train/loss': 0.18093344569206238, 'train/grad_norm_pre_clip': 1.7365700006484985, 'timing/data': 0.0002497709356248379, 'timing/model': 1.2162553339730948, 'learning_rate/action_model': 8.800533134226207e-05, 'learning_rate/action_query_proj': 8.800533134226207e-05, 'learning_rate/base': 2.2001332835565517e-05, 'epoch': 0.2}) +[2026-06-29 18:59:11,028][starVLA.training.train_starvla][INFO] - Step 461, Loss: {'train/loss': 0.012798971496522427, 'train/grad_norm_pre_clip': 3.309008836746216, 'timing/data': 0.00020682113245129585, 'timing/model': 1.2256034209858626, 'learning_rate/action_model': 8.795542520303095e-05, 'learning_rate/action_query_proj': 8.795542520303095e-05, 'learning_rate/base': 2.198885630075774e-05, 'epoch': 0.2}) +[2026-06-29 18:59:43,424][starVLA.training.train_starvla][INFO] - Step 462, Loss: {'train/loss': 0.001746936235576868, 'train/grad_norm_pre_clip': 3.558084726333618, 'timing/data': 0.00018866593018174171, 'timing/model': 1.223188878968358, 'learning_rate/action_model': 8.790543034736237e-05, 'learning_rate/action_query_proj': 8.790543034736237e-05, 'learning_rate/base': 2.197635758684059e-05, 'epoch': 0.2}) +[2026-06-29 19:00:15,760][starVLA.training.train_starvla][INFO] - Step 463, Loss: {'train/loss': 0.020558957010507584, 'train/grad_norm_pre_clip': 5.775419235229492, 'timing/data': 0.0001991551835089922, 'timing/model': 1.2175140930339694, 'learning_rate/action_model': 8.785534689861365e-05, 'learning_rate/action_query_proj': 8.785534689861365e-05, 'learning_rate/base': 2.1963836724653413e-05, 'epoch': 0.2}) +[2026-06-29 19:00:48,125][starVLA.training.train_starvla][INFO] - Step 464, Loss: {'train/loss': 0.004113740287721157, 'train/grad_norm_pre_clip': 1.7181556224822998, 'timing/data': 0.0001899590715765953, 'timing/model': 1.221426370088011, 'learning_rate/action_model': 8.780517498036074e-05, 'learning_rate/action_query_proj': 8.780517498036074e-05, 'learning_rate/base': 2.1951293745090186e-05, 'epoch': 0.2}) +[2026-06-29 19:01:20,492][starVLA.training.train_starvla][INFO] - Step 465, Loss: {'train/loss': 0.16357757151126862, 'train/grad_norm_pre_clip': 3.0836048126220703, 'timing/data': 0.0002859600353986025, 'timing/model': 1.222793284105137, 'learning_rate/action_model': 8.775491471639787e-05, 'learning_rate/action_query_proj': 8.775491471639787e-05, 'learning_rate/base': 2.1938728679099467e-05, 'epoch': 0.2}) +[2026-06-29 19:01:52,871][starVLA.training.train_starvla][INFO] - Step 466, Loss: {'train/loss': 0.12040290236473083, 'train/grad_norm_pre_clip': 1.2709864377975464, 'timing/data': 0.00028811395168304443, 'timing/model': 1.2187692129518837, 'learning_rate/action_model': 8.770456623073721e-05, 'learning_rate/action_query_proj': 8.770456623073721e-05, 'learning_rate/base': 2.1926141557684303e-05, 'epoch': 0.2}) +[2026-06-29 19:02:25,212][starVLA.training.train_starvla][INFO] - Step 467, Loss: {'train/loss': 0.002006649738177657, 'train/grad_norm_pre_clip': 4.364837169647217, 'timing/data': 0.0001966410782188177, 'timing/model': 1.2190035381354392, 'learning_rate/action_model': 8.765412964760866e-05, 'learning_rate/action_query_proj': 8.765412964760866e-05, 'learning_rate/base': 2.1913532411902165e-05, 'epoch': 0.2}) +[2026-06-29 19:02:57,560][starVLA.training.train_starvla][INFO] - Step 468, Loss: {'train/loss': 0.5102329254150391, 'train/grad_norm_pre_clip': 2.364556312561035, 'timing/data': 0.00019217212684452534, 'timing/model': 1.2222697651013732, 'learning_rate/action_model': 8.760360509145949e-05, 'learning_rate/action_query_proj': 8.760360509145949e-05, 'learning_rate/base': 2.1900901272864873e-05, 'epoch': 0.2}) +[2026-06-29 19:03:29,934][starVLA.training.train_starvla][INFO] - Step 469, Loss: {'train/loss': 0.024265047162771225, 'train/grad_norm_pre_clip': 1.73179030418396, 'timing/data': 0.00019519799388945103, 'timing/model': 1.2180031661409885, 'learning_rate/action_model': 8.7552992686954e-05, 'learning_rate/action_query_proj': 8.7552992686954e-05, 'learning_rate/base': 2.18882481717385e-05, 'epoch': 0.2}) +[2026-06-29 19:04:02,276][starVLA.training.train_starvla][INFO] - Step 470, Loss: {'train/loss': 0.4732677638530731, 'train/grad_norm_pre_clip': 2.8672027587890625, 'timing/data': 0.00019711186178028584, 'timing/model': 1.2196455779485404, 'learning_rate/action_model': 8.750229255897328e-05, 'learning_rate/action_query_proj': 8.750229255897328e-05, 'learning_rate/base': 2.187557313974332e-05, 'epoch': 0.2}) +[2026-06-29 19:04:34,630][starVLA.training.train_starvla][INFO] - Step 471, Loss: {'train/loss': 0.2139323651790619, 'train/grad_norm_pre_clip': 2.7056756019592285, 'timing/data': 0.00019944598898291588, 'timing/model': 1.2228755471296608, 'learning_rate/action_model': 8.745150483261484e-05, 'learning_rate/action_query_proj': 8.745150483261484e-05, 'learning_rate/base': 2.186287620815371e-05, 'epoch': 0.2}) +[2026-06-29 19:05:07,052][starVLA.training.train_starvla][INFO] - Step 472, Loss: {'train/loss': 0.022227352485060692, 'train/grad_norm_pre_clip': 1.3430451154708862, 'timing/data': 0.00033466098830103874, 'timing/model': 1.2255675999913365, 'learning_rate/action_model': 8.740062963319236e-05, 'learning_rate/action_query_proj': 8.740062963319236e-05, 'learning_rate/base': 2.185015740829809e-05, 'epoch': 0.2}) +[2026-06-29 19:05:39,464][starVLA.training.train_starvla][INFO] - Step 473, Loss: {'train/loss': 0.00011374526366125792, 'train/grad_norm_pre_clip': 1.3968411684036255, 'timing/data': 0.00018961820751428604, 'timing/model': 1.2225723988376558, 'learning_rate/action_model': 8.734966708623532e-05, 'learning_rate/action_query_proj': 8.734966708623532e-05, 'learning_rate/base': 2.183741677155883e-05, 'epoch': 0.2}) +[2026-06-29 19:06:11,831][starVLA.training.train_starvla][INFO] - Step 474, Loss: {'train/loss': 0.2597000002861023, 'train/grad_norm_pre_clip': 2.8353111743927, 'timing/data': 0.00018984894268214703, 'timing/model': 1.2258359710685909, 'learning_rate/action_model': 8.729861731748877e-05, 'learning_rate/action_query_proj': 8.729861731748877e-05, 'learning_rate/base': 2.1824654329372192e-05, 'epoch': 0.2}) +[2026-06-29 19:06:44,216][starVLA.training.train_starvla][INFO] - Step 475, Loss: {'train/loss': 0.11902689188718796, 'train/grad_norm_pre_clip': 1.397405743598938, 'timing/data': 0.00019470788538455963, 'timing/model': 1.2231350850779563, 'learning_rate/action_model': 8.724748045291291e-05, 'learning_rate/action_query_proj': 8.724748045291291e-05, 'learning_rate/base': 2.181187011322823e-05, 'epoch': 0.2}) +[2026-06-29 19:07:16,606][starVLA.training.train_starvla][INFO] - Step 476, Loss: {'train/loss': 0.03814404085278511, 'train/grad_norm_pre_clip': 1.771594524383545, 'timing/data': 0.0002165988553315401, 'timing/model': 1.22163753118366, 'learning_rate/action_model': 8.71962566186829e-05, 'learning_rate/action_query_proj': 8.71962566186829e-05, 'learning_rate/base': 2.1799064154670724e-05, 'epoch': 0.2}) +[2026-06-29 19:07:48,986][starVLA.training.train_starvla][INFO] - Step 477, Loss: {'train/loss': 0.0020871267188340425, 'train/grad_norm_pre_clip': 5.639251232147217, 'timing/data': 0.00028952606953680515, 'timing/model': 1.2235076618380845, 'learning_rate/action_model': 8.714494594118842e-05, 'learning_rate/action_query_proj': 8.714494594118842e-05, 'learning_rate/base': 2.1786236485297105e-05, 'epoch': 0.21}) +[2026-06-29 19:08:21,362][starVLA.training.train_starvla][INFO] - Step 478, Loss: {'train/loss': 0.08490491658449173, 'train/grad_norm_pre_clip': 3.0380594730377197, 'timing/data': 0.00019896496087312698, 'timing/model': 1.221768220886588, 'learning_rate/action_model': 8.709354854703348e-05, 'learning_rate/action_query_proj': 8.709354854703348e-05, 'learning_rate/base': 2.177338713675837e-05, 'epoch': 0.21}) +[2026-06-29 19:08:53,746][starVLA.training.train_starvla][INFO] - Step 479, Loss: {'train/loss': 0.017769508063793182, 'train/grad_norm_pre_clip': 2.215714931488037, 'timing/data': 0.000191772123798728, 'timing/model': 1.2231036920566112, 'learning_rate/action_model': 8.704206456303607e-05, 'learning_rate/action_query_proj': 8.704206456303607e-05, 'learning_rate/base': 2.1760516140759017e-05, 'epoch': 0.21}) +[2026-06-29 19:09:26,166][starVLA.training.train_starvla][INFO] - Step 480, Loss: {'train/loss': 0.002793461550027132, 'train/grad_norm_pre_clip': 2.396271228790283, 'timing/data': 0.00021291198208928108, 'timing/model': 1.2273434170056134, 'learning_rate/action_model': 8.699049411622776e-05, 'learning_rate/action_query_proj': 8.699049411622776e-05, 'learning_rate/base': 2.174762352905694e-05, 'epoch': 0.21}) +[2026-06-29 19:09:58,572][starVLA.training.train_starvla][INFO] - Step 481, Loss: {'train/loss': 0.026705488562583923, 'train/grad_norm_pre_clip': 2.107691764831543, 'timing/data': 0.00020036892965435982, 'timing/model': 1.2253248668275774, 'learning_rate/action_model': 8.693883733385353e-05, 'learning_rate/action_query_proj': 8.693883733385353e-05, 'learning_rate/base': 2.173470933346338e-05, 'epoch': 0.21}) +[2026-06-29 19:10:31,106][starVLA.training.train_starvla][INFO] - Step 482, Loss: {'train/loss': 0.3190295100212097, 'train/grad_norm_pre_clip': 2.035465955734253, 'timing/data': 0.00018754391930997372, 'timing/model': 1.220238468144089, 'learning_rate/action_model': 8.688709434337134e-05, 'learning_rate/action_query_proj': 8.688709434337134e-05, 'learning_rate/base': 2.1721773585842835e-05, 'epoch': 0.21}) +[2026-06-29 19:11:03,497][starVLA.training.train_starvla][INFO] - Step 483, Loss: {'train/loss': 0.0005781181971542537, 'train/grad_norm_pre_clip': 1.6182048320770264, 'timing/data': 0.00031880196183919907, 'timing/model': 1.220311228185892, 'learning_rate/action_model': 8.683526527245188e-05, 'learning_rate/action_query_proj': 8.683526527245188e-05, 'learning_rate/base': 2.170881631811297e-05, 'epoch': 0.21}) +[2026-06-29 19:11:35,845][starVLA.training.train_starvla][INFO] - Step 484, Loss: {'train/loss': 0.34900790452957153, 'train/grad_norm_pre_clip': 4.18972110748291, 'timing/data': 0.00041225808672606945, 'timing/model': 1.2214003540575504, 'learning_rate/action_model': 8.678335024897824e-05, 'learning_rate/action_query_proj': 8.678335024897824e-05, 'learning_rate/base': 2.169583756224456e-05, 'epoch': 0.21}) +[2026-06-29 19:12:08,216][starVLA.training.train_starvla][INFO] - Step 485, Loss: {'train/loss': 0.021899808198213577, 'train/grad_norm_pre_clip': 1.7647467851638794, 'timing/data': 0.0001943269744515419, 'timing/model': 1.220486890990287, 'learning_rate/action_model': 8.673134940104557e-05, 'learning_rate/action_query_proj': 8.673134940104557e-05, 'learning_rate/base': 2.168283735026139e-05, 'epoch': 0.21}) +[2026-06-29 19:12:40,628][starVLA.training.train_starvla][INFO] - Step 486, Loss: {'train/loss': 0.7830645442008972, 'train/grad_norm_pre_clip': 1.3697618246078491, 'timing/data': 0.00020951591432094574, 'timing/model': 1.2218322311528027, 'learning_rate/action_model': 8.66792628569608e-05, 'learning_rate/action_query_proj': 8.66792628569608e-05, 'learning_rate/base': 2.16698157142402e-05, 'epoch': 0.21}) +[2026-06-29 19:13:13,083][starVLA.training.train_starvla][INFO] - Step 487, Loss: {'train/loss': 0.036287177354097366, 'train/grad_norm_pre_clip': 1.356160044670105, 'timing/data': 0.00032938108779489994, 'timing/model': 1.2317359140142798, 'learning_rate/action_model': 8.662709074524229e-05, 'learning_rate/action_query_proj': 8.662709074524229e-05, 'learning_rate/base': 2.1656772686310572e-05, 'epoch': 0.21}) +[2026-06-29 19:13:45,526][starVLA.training.train_starvla][INFO] - Step 488, Loss: {'train/loss': 0.03712250664830208, 'train/grad_norm_pre_clip': 3.4024879932403564, 'timing/data': 0.00022503500804305077, 'timing/model': 1.2225398879963905, 'learning_rate/action_model': 8.657483319461954e-05, 'learning_rate/action_query_proj': 8.657483319461954e-05, 'learning_rate/base': 2.1643708298654885e-05, 'epoch': 0.21}) +[2026-06-29 19:14:17,929][starVLA.training.train_starvla][INFO] - Step 489, Loss: {'train/loss': 0.22363874316215515, 'train/grad_norm_pre_clip': 1.4716078042984009, 'timing/data': 0.0002228100784122944, 'timing/model': 1.223248956957832, 'learning_rate/action_model': 8.652249033403287e-05, 'learning_rate/action_query_proj': 8.652249033403287e-05, 'learning_rate/base': 2.1630622583508217e-05, 'epoch': 0.21}) +[2026-06-29 19:14:50,322][starVLA.training.train_starvla][INFO] - Step 490, Loss: {'train/loss': 0.028446581214666367, 'train/grad_norm_pre_clip': 2.643881320953369, 'timing/data': 0.00024528312496840954, 'timing/model': 1.2200172990560532, 'learning_rate/action_model': 8.647006229263306e-05, 'learning_rate/action_query_proj': 8.647006229263306e-05, 'learning_rate/base': 2.1617515573158266e-05, 'epoch': 0.21}) +[2026-06-29 19:15:22,726][starVLA.training.train_starvla][INFO] - Step 491, Loss: {'train/loss': 0.08891616016626358, 'train/grad_norm_pre_clip': 2.2308967113494873, 'timing/data': 0.00018948805518448353, 'timing/model': 1.2241360540501773, 'learning_rate/action_model': 8.641754919978115e-05, 'learning_rate/action_query_proj': 8.641754919978115e-05, 'learning_rate/base': 2.1604387299945287e-05, 'epoch': 0.21}) +[2026-06-29 19:15:55,115][starVLA.training.train_starvla][INFO] - Step 492, Loss: {'train/loss': 0.15751111507415771, 'train/grad_norm_pre_clip': 1.2261470556259155, 'timing/data': 0.00019305385649204254, 'timing/model': 1.2238796399906278, 'learning_rate/action_model': 8.63649511850479e-05, 'learning_rate/action_query_proj': 8.63649511850479e-05, 'learning_rate/base': 2.1591237796261974e-05, 'epoch': 0.21}) +[2026-06-29 19:16:27,553][starVLA.training.train_starvla][INFO] - Step 493, Loss: {'train/loss': 0.15433856844902039, 'train/grad_norm_pre_clip': 1.6310410499572754, 'timing/data': 0.00018758396618068218, 'timing/model': 1.2200736380182207, 'learning_rate/action_model': 8.631226837821375e-05, 'learning_rate/action_query_proj': 8.631226837821375e-05, 'learning_rate/base': 2.1578067094553436e-05, 'epoch': 0.21}) +[2026-06-29 19:16:59,904][starVLA.training.train_starvla][INFO] - Step 494, Loss: {'train/loss': 0.12011771649122238, 'train/grad_norm_pre_clip': 3.612459421157837, 'timing/data': 0.00019650091417133808, 'timing/model': 1.2193968708161265, 'learning_rate/action_model': 8.625950090926824e-05, 'learning_rate/action_query_proj': 8.625950090926824e-05, 'learning_rate/base': 2.156487522731706e-05, 'epoch': 0.21}) +[2026-06-29 19:17:32,210][starVLA.training.train_starvla][INFO] - Step 495, Loss: {'train/loss': 0.0003587785176932812, 'train/grad_norm_pre_clip': 2.7055797576904297, 'timing/data': 0.00018581096082925797, 'timing/model': 1.2231639409437776, 'learning_rate/action_model': 8.620664890840986e-05, 'learning_rate/action_query_proj': 8.620664890840986e-05, 'learning_rate/base': 2.1551662227102464e-05, 'epoch': 0.21}) +[2026-06-29 19:18:04,587][starVLA.training.train_starvla][INFO] - Step 496, Loss: {'train/loss': 0.001996331848204136, 'train/grad_norm_pre_clip': 1.5887067317962646, 'timing/data': 0.00023712683469057083, 'timing/model': 1.2187616920564324, 'learning_rate/action_model': 8.615371250604571e-05, 'learning_rate/action_query_proj': 8.615371250604571e-05, 'learning_rate/base': 2.153842812651143e-05, 'epoch': 0.21}) +[2026-06-29 19:18:36,965][starVLA.training.train_starvla][INFO] - Step 497, Loss: {'train/loss': 0.779342770576477, 'train/grad_norm_pre_clip': 1.7529188394546509, 'timing/data': 0.0002939149271696806, 'timing/model': 1.2221087901853025, 'learning_rate/action_model': 8.610069183279105e-05, 'learning_rate/action_query_proj': 8.610069183279105e-05, 'learning_rate/base': 2.1525172958197763e-05, 'epoch': 0.21}) +[2026-06-29 19:19:09,339][starVLA.training.train_starvla][INFO] - Step 498, Loss: {'train/loss': 0.06686610728502274, 'train/grad_norm_pre_clip': 1.7900478839874268, 'timing/data': 0.00018644193187355995, 'timing/model': 1.2230075970292091, 'learning_rate/action_model': 8.604758701946915e-05, 'learning_rate/action_query_proj': 8.604758701946915e-05, 'learning_rate/base': 2.1511896754867288e-05, 'epoch': 0.21}) +[2026-06-29 19:19:41,651][starVLA.training.train_starvla][INFO] - Step 499, Loss: {'train/loss': 0.02843584679067135, 'train/grad_norm_pre_clip': 2.3284261226654053, 'timing/data': 0.00037158187478780746, 'timing/model': 1.2238650862127542, 'learning_rate/action_model': 8.599439819711085e-05, 'learning_rate/action_query_proj': 8.599439819711085e-05, 'learning_rate/base': 2.1498599549277714e-05, 'epoch': 0.21}) +[2026-06-29 19:20:14,030][starVLA.training.train_starvla][INFO] - Starting held-out action loss eval at step 500 +[2026-06-29 19:21:10,526][starVLA.training.train_starvla][INFO] - Finished held-out action loss eval at step 500 +[2026-06-29 19:21:10,527][starVLA.training.train_starvla][INFO] - Starting held-out action classification eval at step 500 +[2026-06-29 19:21:10,527][starVLA.training.train_starvla][INFO] - Finished held-out action classification eval at step 500 +[2026-06-29 19:21:10,676][root][INFO] - gcc -pthread -B /venv/starvla_wanoft/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall -fPIC -fno-merge-constants -O2 -isystem /venv/starvla_wanoft/include -fPIC -fno-merge-constants -O2 -isystem /venv/starvla_wanoft/include -fPIC -c /tmp/tmp_fs4535b/test.c -o /tmp/tmp_fs4535b/test.o +[2026-06-29 19:21:10,700][root][INFO] - gcc -pthread -B /venv/starvla_wanoft/compiler_compat /tmp/tmp_fs4535b/test.o -laio -o /tmp/tmp_fs4535b/a.out +[2026-06-29 19:21:11,358][root][INFO] - gcc -pthread -B /venv/starvla_wanoft/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall -fPIC -fno-merge-constants -O2 -isystem /venv/starvla_wanoft/include -fPIC -fno-merge-constants -O2 -isystem /venv/starvla_wanoft/include -fPIC -c /tmp/tmpk5nlqv0y/test.c -o /tmp/tmpk5nlqv0y/test.o +[2026-06-29 19:21:11,365][root][INFO] - gcc -pthread -B /venv/starvla_wanoft/compiler_compat /tmp/tmpk5nlqv0y/test.o -L/usr/local/cuda -L/usr/local/cuda/lib64 -lcufile -o /tmp/tmpk5nlqv0y/a.out +[2026-06-29 19:21:11,390][root][INFO] - gcc -pthread -B /venv/starvla_wanoft/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall -fPIC -fno-merge-constants -O2 -isystem /venv/starvla_wanoft/include -fPIC -fno-merge-constants -O2 -isystem /venv/starvla_wanoft/include -fPIC -c /tmp/tmpvc53yug7/test.c -o /tmp/tmpvc53yug7/test.o +[2026-06-29 19:21:11,396][root][INFO] - gcc -pthread -B /venv/starvla_wanoft/compiler_compat /tmp/tmpvc53yug7/test.o -laio -o /tmp/tmpvc53yug7/a.out +[2026-06-29 20:03:00,368][starVLA.training.rl_games.checkpoint_sync][INFO] - HF checkpoint sync disabled; skipping eval result upload for step 500 +[2026-06-29 20:03:00,369][starVLA.training.train_starvla][INFO] - Step 500, Loss: {'train/loss': 0.01963452622294426, 'train/grad_norm_pre_clip': 2.3429348468780518, 'eval/action_loss/seconds': 56.493678187951446, 'eval/action_loss/samples': 200.0, 'eval/loss': 0.05574990436434746, 'eval/latency_1/loss': 0.05574990436434746, 'eval/flappy/loss': 0.05574990436434746, 'eval/flappy/latency_1/loss': 0.05574990436434746, 'rl_games_eval/mid_train/total_episodes': 5.0, 'rl_games_eval/mid_train/mean_reward': 382.18000000001996, 'rl_games_eval/mid_train/mean_length': 3096.8, 'rl_games_eval/mid_train/std_reward': 124.8400000000113, 'rl_games_eval/mid_train/std_length': 1006.4000000000002, 'rl_games_eval/mid_train/macro_mean_reward': 382.18000000001996, 'rl_games_eval/mid_train/macro_mean_length': 3096.8, 'rl_games_eval/mid_train/task_count': 1.0, 'rl_games_eval/mid_train/flappy__latency_1/mean_reward': 382.18000000001996, 'rl_games_eval/mid_train/flappy__latency_1/mean_length': 3096.8, 'rl_games_eval/mid_train/flappy__latency_1/std_reward': 124.8400000000113, 'rl_games_eval/mid_train/flappy__latency_1/std_length': 1006.4000000000002, 'rl_games_eval/mid_train/flappy__latency_1/num_episodes': 5.0, 'checkpoint/current_eval_score': 382.18000000001996, 'checkpoint/best_score': -inf, 'checkpoint/best_step': 0.0, 'checkpoint/is_new_best': 0.0, 'timing/data': 0.00019285408779978752, 'timing/model': 1.2231049619149417, 'learning_rate/action_model': 8.594112549695428e-05, 'learning_rate/action_query_proj': 8.594112549695428e-05, 'learning_rate/base': 2.148528137423857e-05, 'epoch': 0.22}) +[2026-06-29 20:03:33,083][starVLA.training.train_starvla][INFO] - Step 501, Loss: {'train/loss': 0.222431018948555, 'train/grad_norm_pre_clip': 1.4084686040878296, 'timing/data': 0.00031048478558659554, 'timing/model': 1.2298811718355864, 'learning_rate/action_model': 8.588776905044454e-05, 'learning_rate/action_query_proj': 8.588776905044454e-05, 'learning_rate/base': 2.1471942262611136e-05, 'epoch': 0.22}) +[2026-06-29 20:04:05,753][starVLA.training.train_starvla][INFO] - Step 502, Loss: {'train/loss': 0.007832936942577362, 'train/grad_norm_pre_clip': 1.488629937171936, 'timing/data': 0.0002567628398537636, 'timing/model': 1.2341801680158824, 'learning_rate/action_model': 8.583432898923335e-05, 'learning_rate/action_query_proj': 8.583432898923335e-05, 'learning_rate/base': 2.1458582247308337e-05, 'epoch': 0.22}) +[2026-06-29 20:04:38,434][starVLA.training.train_starvla][INFO] - Step 503, Loss: {'train/loss': 0.0015820072731003165, 'train/grad_norm_pre_clip': 2.1646740436553955, 'timing/data': 0.0003210739232599735, 'timing/model': 1.227887982968241, 'learning_rate/action_model': 8.578080544517876e-05, 'learning_rate/action_query_proj': 8.578080544517876e-05, 'learning_rate/base': 2.144520136129469e-05, 'epoch': 0.22}) +[2026-06-29 20:05:11,089][starVLA.training.train_starvla][INFO] - Step 504, Loss: {'train/loss': 0.01841558888554573, 'train/grad_norm_pre_clip': 1.8563023805618286, 'timing/data': 0.00028713117353618145, 'timing/model': 1.2334294610191137, 'learning_rate/action_model': 8.572719855034477e-05, 'learning_rate/action_query_proj': 8.572719855034477e-05, 'learning_rate/base': 2.1431799637586193e-05, 'epoch': 0.22}) +[2026-06-29 20:05:43,794][starVLA.training.train_starvla][INFO] - Step 505, Loss: {'train/loss': 0.0964033231139183, 'train/grad_norm_pre_clip': 3.9582912921905518, 'timing/data': 0.0002951761707663536, 'timing/model': 1.2338783768936992, 'learning_rate/action_model': 8.567350843700108e-05, 'learning_rate/action_query_proj': 8.567350843700108e-05, 'learning_rate/base': 2.141837710925027e-05, 'epoch': 0.22}) +[2026-06-29 20:06:16,488][starVLA.training.train_starvla][INFO] - Step 506, Loss: {'train/loss': 0.03453386202454567, 'train/grad_norm_pre_clip': 1.3826649188995361, 'timing/data': 0.00029192003421485424, 'timing/model': 1.2328423159196973, 'learning_rate/action_model': 8.56197352376227e-05, 'learning_rate/action_query_proj': 8.56197352376227e-05, 'learning_rate/base': 2.1404933809405676e-05, 'epoch': 0.22}) +[2026-06-29 20:06:49,166][starVLA.training.train_starvla][INFO] - Step 507, Loss: {'train/loss': 0.0012395347002893686, 'train/grad_norm_pre_clip': 2.957737445831299, 'timing/data': 0.00026562088169157505, 'timing/model': 1.2270295759662986, 'learning_rate/action_model': 8.556587908488969e-05, 'learning_rate/action_query_proj': 8.556587908488969e-05, 'learning_rate/base': 2.1391469771222423e-05, 'epoch': 0.22}) +[2026-06-29 20:07:21,819][starVLA.training.train_starvla][INFO] - Step 508, Loss: {'train/loss': 0.15091168880462646, 'train/grad_norm_pre_clip': 3.038438081741333, 'timing/data': 0.0001936350017786026, 'timing/model': 1.2295088320970535, 'learning_rate/action_model': 8.55119401116867e-05, 'learning_rate/action_query_proj': 8.55119401116867e-05, 'learning_rate/base': 2.1377985027921675e-05, 'epoch': 0.22}) +[2026-06-29 20:07:54,500][starVLA.training.train_starvla][INFO] - Step 509, Loss: {'train/loss': 0.02023754082620144, 'train/grad_norm_pre_clip': 2.0725154876708984, 'timing/data': 0.00019421614706516266, 'timing/model': 1.23524412792176, 'learning_rate/action_model': 8.54579184511028e-05, 'learning_rate/action_query_proj': 8.54579184511028e-05, 'learning_rate/base': 2.13644796127757e-05, 'epoch': 0.22}) +[2026-06-29 20:08:27,169][starVLA.training.train_starvla][INFO] - Step 510, Loss: {'train/loss': 0.011804783716797829, 'train/grad_norm_pre_clip': 3.466789484024048, 'timing/data': 0.000194215914234519, 'timing/model': 1.2371401228010654, 'learning_rate/action_model': 8.54038142364311e-05, 'learning_rate/action_query_proj': 8.54038142364311e-05, 'learning_rate/base': 2.1350953559107774e-05, 'epoch': 0.22}) +[2026-06-29 20:08:59,894][starVLA.training.train_starvla][INFO] - Step 511, Loss: {'train/loss': 0.7976435422897339, 'train/grad_norm_pre_clip': 2.9782536029815674, 'timing/data': 0.00025655399076640606, 'timing/model': 1.2344648959115148, 'learning_rate/action_model': 8.534962760116833e-05, 'learning_rate/action_query_proj': 8.534962760116833e-05, 'learning_rate/base': 2.133740690029208e-05, 'epoch': 0.22}) +[2026-06-29 20:09:32,584][starVLA.training.train_starvla][INFO] - Step 512, Loss: {'train/loss': 0.0008798963390290737, 'train/grad_norm_pre_clip': 2.0597331523895264, 'timing/data': 0.00020007696002721786, 'timing/model': 1.2294479708652943, 'learning_rate/action_model': 8.529535867901463e-05, 'learning_rate/action_query_proj': 8.529535867901463e-05, 'learning_rate/base': 2.1323839669753658e-05, 'epoch': 0.22}) +[2026-06-29 20:10:05,302][starVLA.training.train_starvla][INFO] - Step 513, Loss: {'train/loss': 0.0061391317285597324, 'train/grad_norm_pre_clip': 3.8048439025878906, 'timing/data': 0.0001970010343939066, 'timing/model': 1.2365507299546152, 'learning_rate/action_model': 8.524100760387321e-05, 'learning_rate/action_query_proj': 8.524100760387321e-05, 'learning_rate/base': 2.1310251900968303e-05, 'epoch': 0.22}) +[2026-06-29 20:10:38,001][starVLA.training.train_starvla][INFO] - Step 514, Loss: {'train/loss': 0.004652935080230236, 'train/grad_norm_pre_clip': 2.261592149734497, 'timing/data': 0.0001937949564307928, 'timing/model': 1.2305975079070777, 'learning_rate/action_model': 8.518657450984991e-05, 'learning_rate/action_query_proj': 8.518657450984991e-05, 'learning_rate/base': 2.1296643627462478e-05, 'epoch': 0.22}) +[2026-06-29 20:11:10,751][starVLA.training.train_starvla][INFO] - Step 515, Loss: {'train/loss': 0.00022138736676424742, 'train/grad_norm_pre_clip': 1.5187194347381592, 'timing/data': 0.0002903880085796118, 'timing/model': 1.2312513650394976, 'learning_rate/action_model': 8.5132059531253e-05, 'learning_rate/action_query_proj': 8.5132059531253e-05, 'learning_rate/base': 2.128301488281325e-05, 'epoch': 0.22}) +[2026-06-29 20:11:43,451][starVLA.training.train_starvla][INFO] - Step 516, Loss: {'train/loss': 0.0004852190613746643, 'train/grad_norm_pre_clip': 1.458840250968933, 'timing/data': 0.00020180991850793362, 'timing/model': 1.2341261520050466, 'learning_rate/action_model': 8.507746280259273e-05, 'learning_rate/action_query_proj': 8.507746280259273e-05, 'learning_rate/base': 2.1269365700648182e-05, 'epoch': 0.22}) +[2026-06-29 20:12:16,151][starVLA.training.train_starvla][INFO] - Step 517, Loss: {'train/loss': 0.29764309525489807, 'train/grad_norm_pre_clip': 7.3526506423950195, 'timing/data': 0.00019341404549777508, 'timing/model': 1.230484542902559, 'learning_rate/action_model': 8.502278445858114e-05, 'learning_rate/action_query_proj': 8.502278445858114e-05, 'learning_rate/base': 2.1255696114645286e-05, 'epoch': 0.22}) +[2026-06-29 20:12:48,830][starVLA.training.train_starvla][INFO] - Step 518, Loss: {'train/loss': 0.03711765632033348, 'train/grad_norm_pre_clip': 2.902216911315918, 'timing/data': 0.00022610696032643318, 'timing/model': 1.2219906128011644, 'learning_rate/action_model': 8.49680246341316e-05, 'learning_rate/action_query_proj': 8.49680246341316e-05, 'learning_rate/base': 2.12420061585329e-05, 'epoch': 0.22}) +[2026-06-29 20:13:21,540][starVLA.training.train_starvla][INFO] - Step 519, Loss: {'train/loss': 0.0005561785656027496, 'train/grad_norm_pre_clip': 3.047647476196289, 'timing/data': 0.0003223579842597246, 'timing/model': 1.2305396050214767, 'learning_rate/action_model': 8.491318346435852e-05, 'learning_rate/action_query_proj': 8.491318346435852e-05, 'learning_rate/base': 2.122829586608963e-05, 'epoch': 0.22}) +[2026-06-29 20:13:54,269][starVLA.training.train_starvla][INFO] - Step 520, Loss: {'train/loss': 0.005724807269871235, 'train/grad_norm_pre_clip': 4.596505641937256, 'timing/data': 0.00019929604604840279, 'timing/model': 1.2346657968591899, 'learning_rate/action_model': 8.485826108457706e-05, 'learning_rate/action_query_proj': 8.485826108457706e-05, 'learning_rate/base': 2.1214565271144265e-05, 'epoch': 0.22}) +[2026-06-29 20:14:26,949][starVLA.training.train_starvla][INFO] - Step 521, Loss: {'train/loss': 0.0025830401573330164, 'train/grad_norm_pre_clip': 2.905769109725952, 'timing/data': 0.00033467100001871586, 'timing/model': 1.2333513589110225, 'learning_rate/action_model': 8.480325763030272e-05, 'learning_rate/action_query_proj': 8.480325763030272e-05, 'learning_rate/base': 2.120081440757568e-05, 'epoch': 0.22}) +[2026-06-29 20:14:59,819][starVLA.training.train_starvla][INFO] - Step 522, Loss: {'train/loss': 0.01254261378198862, 'train/grad_norm_pre_clip': 3.1078779697418213, 'timing/data': 0.000264317961409688, 'timing/model': 1.2292283028364182, 'learning_rate/action_model': 8.474817323725104e-05, 'learning_rate/action_query_proj': 8.474817323725104e-05, 'learning_rate/base': 2.118704330931276e-05, 'epoch': 0.22}) +[2026-06-29 20:15:32,518][starVLA.training.train_starvla][INFO] - Step 523, Loss: {'train/loss': 0.1597573608160019, 'train/grad_norm_pre_clip': 3.5717968940734863, 'timing/data': 0.00018725311383605003, 'timing/model': 1.232089360943064, 'learning_rate/action_model': 8.469300804133729e-05, 'learning_rate/action_query_proj': 8.469300804133729e-05, 'learning_rate/base': 2.1173252010334322e-05, 'epoch': 0.23}) +[2026-06-29 20:16:05,202][starVLA.training.train_starvla][INFO] - Step 524, Loss: {'train/loss': 0.00022480907500721514, 'train/grad_norm_pre_clip': 1.5502859354019165, 'timing/data': 0.00019189203158020973, 'timing/model': 1.2305081721860915, 'learning_rate/action_model': 8.463776217867613e-05, 'learning_rate/action_query_proj': 8.463776217867613e-05, 'learning_rate/base': 2.115944054466903e-05, 'epoch': 0.23}) +[2026-06-29 20:16:37,890][starVLA.training.train_starvla][INFO] - Step 525, Loss: {'train/loss': 0.06583913415670395, 'train/grad_norm_pre_clip': 2.594162940979004, 'timing/data': 0.00025962013751268387, 'timing/model': 1.230709180003032, 'learning_rate/action_model': 8.458243578558122e-05, 'learning_rate/action_query_proj': 8.458243578558122e-05, 'learning_rate/base': 2.1145608946395304e-05, 'epoch': 0.23}) +[2026-06-29 20:17:10,559][starVLA.training.train_starvla][INFO] - Step 526, Loss: {'train/loss': 0.06064716354012489, 'train/grad_norm_pre_clip': 1.824040412902832, 'timing/data': 0.00019786320626735687, 'timing/model': 1.231228734832257, 'learning_rate/action_model': 8.452702899856491e-05, 'learning_rate/action_query_proj': 8.452702899856491e-05, 'learning_rate/base': 2.113175724964123e-05, 'epoch': 0.23}) +[2026-06-29 20:17:43,272][starVLA.training.train_starvla][INFO] - Step 527, Loss: {'train/loss': 0.033834561705589294, 'train/grad_norm_pre_clip': 2.609739065170288, 'timing/data': 0.00019608903676271439, 'timing/model': 1.2307932879775763, 'learning_rate/action_model': 8.447154195433799e-05, 'learning_rate/action_query_proj': 8.447154195433799e-05, 'learning_rate/base': 2.1117885488584497e-05, 'epoch': 0.23}) +[2026-06-29 20:18:16,014][starVLA.training.train_starvla][INFO] - Step 528, Loss: {'train/loss': 0.3397215008735657, 'train/grad_norm_pre_clip': 2.4609899520874023, 'timing/data': 0.0004826399963349104, 'timing/model': 1.226541514042765, 'learning_rate/action_model': 8.441597478980917e-05, 'learning_rate/action_query_proj': 8.441597478980917e-05, 'learning_rate/base': 2.1103993697452293e-05, 'epoch': 0.23}) +[2026-06-29 20:18:48,734][starVLA.training.train_starvla][INFO] - Step 529, Loss: {'train/loss': 0.1436564326286316, 'train/grad_norm_pre_clip': 2.5428733825683594, 'timing/data': 0.00047469488345086575, 'timing/model': 1.2318440210074186, 'learning_rate/action_model': 8.436032764208494e-05, 'learning_rate/action_query_proj': 8.436032764208494e-05, 'learning_rate/base': 2.1090081910521236e-05, 'epoch': 0.23}) +[2026-06-29 20:19:21,393][starVLA.training.train_starvla][INFO] - Step 530, Loss: {'train/loss': 0.040892817080020905, 'train/grad_norm_pre_clip': 4.303030967712402, 'timing/data': 0.00023928098380565643, 'timing/model': 1.2322319308295846, 'learning_rate/action_model': 8.430460064846912e-05, 'learning_rate/action_query_proj': 8.430460064846912e-05, 'learning_rate/base': 2.107615016211728e-05, 'epoch': 0.23}) +[2026-06-29 20:19:54,110][starVLA.training.train_starvla][INFO] - Step 531, Loss: {'train/loss': 0.03337720036506653, 'train/grad_norm_pre_clip': 2.0292725563049316, 'timing/data': 0.00020339293405413628, 'timing/model': 1.2313657139893621, 'learning_rate/action_model': 8.42487939464625e-05, 'learning_rate/action_query_proj': 8.42487939464625e-05, 'learning_rate/base': 2.1062198486615625e-05, 'epoch': 0.23}) +[2026-06-29 20:20:26,830][starVLA.training.train_starvla][INFO] - Step 532, Loss: {'train/loss': 0.619495153427124, 'train/grad_norm_pre_clip': 2.834549903869629, 'timing/data': 0.00019687111489474773, 'timing/model': 1.2318490101024508, 'learning_rate/action_model': 8.419290767376255e-05, 'learning_rate/action_query_proj': 8.419290767376255e-05, 'learning_rate/base': 2.1048226918440637e-05, 'epoch': 0.23}) +[2026-06-29 20:20:59,499][starVLA.training.train_starvla][INFO] - Step 533, Loss: {'train/loss': 0.06607622653245926, 'train/grad_norm_pre_clip': 3.333893299102783, 'timing/data': 0.0002559819258749485, 'timing/model': 1.2348316968418658, 'learning_rate/action_model': 8.413694196826315e-05, 'learning_rate/action_query_proj': 8.413694196826315e-05, 'learning_rate/base': 2.1034235492065788e-05, 'epoch': 0.23}) +[2026-06-29 20:21:32,217][starVLA.training.train_starvla][INFO] - Step 534, Loss: {'train/loss': 0.15727372467517853, 'train/grad_norm_pre_clip': 2.6520395278930664, 'timing/data': 0.00019177095964550972, 'timing/model': 1.2314939759671688, 'learning_rate/action_model': 8.408089696805407e-05, 'learning_rate/action_query_proj': 8.408089696805407e-05, 'learning_rate/base': 2.1020224242013517e-05, 'epoch': 0.23}) +[2026-06-29 20:22:04,890][starVLA.training.train_starvla][INFO] - Step 535, Loss: {'train/loss': 0.002397643169388175, 'train/grad_norm_pre_clip': 2.7705323696136475, 'timing/data': 0.00020229211077094078, 'timing/model': 1.2281606220640242, 'learning_rate/action_model': 8.40247728114208e-05, 'learning_rate/action_query_proj': 8.40247728114208e-05, 'learning_rate/base': 2.10061932028552e-05, 'epoch': 0.23}) +[2026-06-29 20:22:37,597][starVLA.training.train_starvla][INFO] - Step 536, Loss: {'train/loss': 0.0006687335553579032, 'train/grad_norm_pre_clip': 4.583194732666016, 'timing/data': 0.00019352510571479797, 'timing/model': 1.2309893348719925, 'learning_rate/action_model': 8.396856963684409e-05, 'learning_rate/action_query_proj': 8.396856963684409e-05, 'learning_rate/base': 2.0992142409211022e-05, 'epoch': 0.23}) +[2026-06-29 20:23:10,308][starVLA.training.train_starvla][INFO] - Step 537, Loss: {'train/loss': 0.04808609560132027, 'train/grad_norm_pre_clip': 1.7123512029647827, 'timing/data': 0.00019924482330679893, 'timing/model': 1.232098723994568, 'learning_rate/action_model': 8.39122875829997e-05, 'learning_rate/action_query_proj': 8.39122875829997e-05, 'learning_rate/base': 2.0978071895749924e-05, 'epoch': 0.23}) +[2026-06-29 20:23:43,047][starVLA.training.train_starvla][INFO] - Step 538, Loss: {'train/loss': 0.3023553788661957, 'train/grad_norm_pre_clip': 2.3073863983154297, 'timing/data': 0.00019159214571118355, 'timing/model': 1.2379852798767388, 'learning_rate/action_model': 8.385592678875802e-05, 'learning_rate/action_query_proj': 8.385592678875802e-05, 'learning_rate/base': 2.0963981697189505e-05, 'epoch': 0.23}) +[2026-06-29 20:24:15,762][starVLA.training.train_starvla][INFO] - Step 539, Loss: {'train/loss': 0.31946226954460144, 'train/grad_norm_pre_clip': 5.378712177276611, 'timing/data': 0.0002571151126176119, 'timing/model': 1.2340876469388604, 'learning_rate/action_model': 8.379948739318368e-05, 'learning_rate/action_query_proj': 8.379948739318368e-05, 'learning_rate/base': 2.094987184829592e-05, 'epoch': 0.23}) +[2026-06-29 20:24:48,405][starVLA.training.train_starvla][INFO] - Step 540, Loss: {'train/loss': 0.017410453408956528, 'train/grad_norm_pre_clip': 5.013672351837158, 'timing/data': 0.00019770395010709763, 'timing/model': 1.233103743987158, 'learning_rate/action_model': 8.374296953553528e-05, 'learning_rate/action_query_proj': 8.374296953553528e-05, 'learning_rate/base': 2.093574238388382e-05, 'epoch': 0.23}) +[2026-06-29 20:25:21,069][starVLA.training.train_starvla][INFO] - Step 541, Loss: {'train/loss': 0.44479623436927795, 'train/grad_norm_pre_clip': 3.516209602355957, 'timing/data': 0.00019783410243690014, 'timing/model': 1.2296599161345512, 'learning_rate/action_model': 8.368637335526505e-05, 'learning_rate/action_query_proj': 8.368637335526505e-05, 'learning_rate/base': 2.0921593338816263e-05, 'epoch': 0.23}) +[2026-06-29 20:25:53,759][starVLA.training.train_starvla][INFO] - Step 542, Loss: {'train/loss': 0.00663028797134757, 'train/grad_norm_pre_clip': 4.490411281585693, 'timing/data': 0.00021575810387730598, 'timing/model': 1.2289174629841, 'learning_rate/action_model': 8.362969899201842e-05, 'learning_rate/action_query_proj': 8.362969899201842e-05, 'learning_rate/base': 2.0907424748004606e-05, 'epoch': 0.23}) +[2026-06-29 20:26:26,427][starVLA.training.train_starvla][INFO] - Step 543, Loss: {'train/loss': 0.5388761758804321, 'train/grad_norm_pre_clip': 1.2814451456069946, 'timing/data': 0.0002504128497093916, 'timing/model': 1.2316109710372984, 'learning_rate/action_model': 8.357294658563373e-05, 'learning_rate/action_query_proj': 8.357294658563373e-05, 'learning_rate/base': 2.0893236646408434e-05, 'epoch': 0.23}) +[2026-06-29 20:26:59,146][starVLA.training.train_starvla][INFO] - Step 544, Loss: {'train/loss': 0.007399622816592455, 'train/grad_norm_pre_clip': 1.6048858165740967, 'timing/data': 0.00024320906959474087, 'timing/model': 1.2229876229539514, 'learning_rate/action_model': 8.35161162761419e-05, 'learning_rate/action_query_proj': 8.35161162761419e-05, 'learning_rate/base': 2.0879029069035476e-05, 'epoch': 0.23}) +[2026-06-29 20:27:31,855][starVLA.training.train_starvla][INFO] - Step 545, Loss: {'train/loss': 0.4110313653945923, 'train/grad_norm_pre_clip': 1.775488018989563, 'timing/data': 0.00028358493000268936, 'timing/model': 1.2297939429990947, 'learning_rate/action_model': 8.34592082037661e-05, 'learning_rate/action_query_proj': 8.34592082037661e-05, 'learning_rate/base': 2.0864802050941526e-05, 'epoch': 0.23}) +[2026-06-29 20:28:04,551][starVLA.training.train_starvla][INFO] - Step 546, Loss: {'train/loss': 0.0712381973862648, 'train/grad_norm_pre_clip': 1.527382254600525, 'timing/data': 0.00019064010120928288, 'timing/model': 1.2346398730296642, 'learning_rate/action_model': 8.340222250892133e-05, 'learning_rate/action_query_proj': 8.340222250892133e-05, 'learning_rate/base': 2.0850555627230334e-05, 'epoch': 0.23}) +[2026-06-29 20:28:37,259][starVLA.training.train_starvla][INFO] - Step 547, Loss: {'train/loss': 0.016576778143644333, 'train/grad_norm_pre_clip': 1.594599723815918, 'timing/data': 0.0002547518815845251, 'timing/model': 1.2344290928449482, 'learning_rate/action_model': 8.334515933221411e-05, 'learning_rate/action_query_proj': 8.334515933221411e-05, 'learning_rate/base': 2.0836289833053528e-05, 'epoch': 0.24}) +[2026-06-29 20:29:09,923][starVLA.training.train_starvla][INFO] - Step 548, Loss: {'train/loss': 0.15330661833286285, 'train/grad_norm_pre_clip': 2.084989309310913, 'timing/data': 0.00041156704537570477, 'timing/model': 1.2342228160705417, 'learning_rate/action_model': 8.328801881444218e-05, 'learning_rate/action_query_proj': 8.328801881444218e-05, 'learning_rate/base': 2.0822004703610546e-05, 'epoch': 0.24}) +[2026-06-29 20:29:42,633][starVLA.training.train_starvla][INFO] - Step 549, Loss: {'train/loss': 0.009696993045508862, 'train/grad_norm_pre_clip': 4.99651575088501, 'timing/data': 0.00027777417562901974, 'timing/model': 1.2298629961442202, 'learning_rate/action_model': 8.323080109659406e-05, 'learning_rate/action_query_proj': 8.323080109659406e-05, 'learning_rate/base': 2.0807700274148516e-05, 'epoch': 0.24}) +[2026-06-29 20:30:15,297][starVLA.training.train_starvla][INFO] - Step 550, Loss: {'train/loss': 8.841676026349887e-05, 'train/grad_norm_pre_clip': 2.262723684310913, 'timing/data': 0.00019647111184895039, 'timing/model': 1.2255020858719945, 'learning_rate/action_model': 8.31735063198488e-05, 'learning_rate/action_query_proj': 8.31735063198488e-05, 'learning_rate/base': 2.07933765799622e-05, 'epoch': 0.24}) +[2026-06-29 20:30:47,993][starVLA.training.train_starvla][INFO] - Step 551, Loss: {'train/loss': 0.6908290982246399, 'train/grad_norm_pre_clip': 3.0310397148132324, 'timing/data': 0.0002545600291341543, 'timing/model': 1.2279782199766487, 'learning_rate/action_model': 8.311613462557559e-05, 'learning_rate/action_query_proj': 8.311613462557559e-05, 'learning_rate/base': 2.0779033656393897e-05, 'epoch': 0.24}) +[2026-06-29 20:31:20,687][starVLA.training.train_starvla][INFO] - Step 552, Loss: {'train/loss': 0.004721482750028372, 'train/grad_norm_pre_clip': 2.0096375942230225, 'timing/data': 0.0002011498436331749, 'timing/model': 1.2318019771482795, 'learning_rate/action_model': 8.305868615533335e-05, 'learning_rate/action_query_proj': 8.305868615533335e-05, 'learning_rate/base': 2.0764671538833337e-05, 'epoch': 0.24}) +[2026-06-29 20:31:53,422][starVLA.training.train_starvla][INFO] - Step 553, Loss: {'train/loss': 0.07723414152860641, 'train/grad_norm_pre_clip': 3.830225944519043, 'timing/data': 0.00019428692758083344, 'timing/model': 1.2286107880063355, 'learning_rate/action_model': 8.300116105087043e-05, 'learning_rate/action_query_proj': 8.300116105087043e-05, 'learning_rate/base': 2.075029026271761e-05, 'epoch': 0.24}) +[2026-06-29 20:32:26,124][starVLA.training.train_starvla][INFO] - Step 554, Loss: {'train/loss': 0.02176791802048683, 'train/grad_norm_pre_clip': 4.971926689147949, 'timing/data': 0.0003400419373065233, 'timing/model': 1.2310574930161238, 'learning_rate/action_model': 8.294355945412439e-05, 'learning_rate/action_query_proj': 8.294355945412439e-05, 'learning_rate/base': 2.0735889863531098e-05, 'epoch': 0.24}) +[2026-06-29 20:32:58,797][starVLA.training.train_starvla][INFO] - Step 555, Loss: {'train/loss': 0.21387986838817596, 'train/grad_norm_pre_clip': 6.125032424926758, 'timing/data': 0.0002010201569646597, 'timing/model': 1.2291839299723506, 'learning_rate/action_model': 8.28858815072214e-05, 'learning_rate/action_query_proj': 8.28858815072214e-05, 'learning_rate/base': 2.072147037680535e-05, 'epoch': 0.24}) +[2026-06-29 20:33:31,476][starVLA.training.train_starvla][INFO] - Step 556, Loss: {'train/loss': 0.00020192319061607122, 'train/grad_norm_pre_clip': 3.0811400413513184, 'timing/data': 0.00033691711723804474, 'timing/model': 1.2316549560055137, 'learning_rate/action_model': 8.282812735247606e-05, 'learning_rate/action_query_proj': 8.282812735247606e-05, 'learning_rate/base': 2.0707031838119015e-05, 'epoch': 0.24}) +[2026-06-29 20:34:04,165][starVLA.training.train_starvla][INFO] - Step 557, Loss: {'train/loss': 0.0020068564917892218, 'train/grad_norm_pre_clip': 3.224299669265747, 'timing/data': 0.00019870512187480927, 'timing/model': 1.2311261559370905, 'learning_rate/action_model': 8.2770297132391e-05, 'learning_rate/action_query_proj': 8.2770297132391e-05, 'learning_rate/base': 2.069257428309775e-05, 'epoch': 0.24}) +[2026-06-29 20:34:36,821][starVLA.training.train_starvla][INFO] - Step 558, Loss: {'train/loss': 0.0003622857038863003, 'train/grad_norm_pre_clip': 3.2698097229003906, 'timing/data': 0.0003668928984552622, 'timing/model': 1.2323869720567018, 'learning_rate/action_model': 8.271239098965656e-05, 'learning_rate/action_query_proj': 8.271239098965656e-05, 'learning_rate/base': 2.067809774741414e-05, 'epoch': 0.24}) +[2026-06-29 20:35:09,523][starVLA.training.train_starvla][INFO] - Step 559, Loss: {'train/loss': 0.0028119708877056837, 'train/grad_norm_pre_clip': 2.929347276687622, 'timing/data': 0.00019900593906641006, 'timing/model': 1.2293293629772961, 'learning_rate/action_model': 8.265440906715039e-05, 'learning_rate/action_query_proj': 8.265440906715039e-05, 'learning_rate/base': 2.0663602266787598e-05, 'epoch': 0.24}) +[2026-06-29 20:35:42,212][starVLA.training.train_starvla][INFO] - Step 560, Loss: {'train/loss': 0.0003524345811456442, 'train/grad_norm_pre_clip': 3.3396518230438232, 'timing/data': 0.00027027097530663013, 'timing/model': 1.2296174550428987, 'learning_rate/action_model': 8.25963515079371e-05, 'learning_rate/action_query_proj': 8.25963515079371e-05, 'learning_rate/base': 2.0649087876984276e-05, 'epoch': 0.24}) +[2026-06-29 20:36:14,915][starVLA.training.train_starvla][INFO] - Step 561, Loss: {'train/loss': 0.016118939965963364, 'train/grad_norm_pre_clip': 1.2183092832565308, 'timing/data': 0.0003252739552408457, 'timing/model': 1.2302290019579232, 'learning_rate/action_model': 8.253821845526798e-05, 'learning_rate/action_query_proj': 8.253821845526798e-05, 'learning_rate/base': 2.0634554613816996e-05, 'epoch': 0.24}) +[2026-06-29 20:36:47,519][starVLA.training.train_starvla][INFO] - Step 562, Loss: {'train/loss': 0.004208284430205822, 'train/grad_norm_pre_clip': 1.9269578456878662, 'timing/data': 0.00019155209884047508, 'timing/model': 1.229726605117321, 'learning_rate/action_model': 8.248001005258052e-05, 'learning_rate/action_query_proj': 8.248001005258052e-05, 'learning_rate/base': 2.062000251314513e-05, 'epoch': 0.24}) +[2026-06-29 20:37:20,236][starVLA.training.train_starvla][INFO] - Step 563, Loss: {'train/loss': 0.3313087522983551, 'train/grad_norm_pre_clip': 2.526599645614624, 'timing/data': 0.0003059979062527418, 'timing/model': 1.2298236519563943, 'learning_rate/action_model': 8.242172644349819e-05, 'learning_rate/action_query_proj': 8.242172644349819e-05, 'learning_rate/base': 2.0605431610874546e-05, 'epoch': 0.24}) +[2026-06-29 20:37:52,953][starVLA.training.train_starvla][INFO] - Step 564, Loss: {'train/loss': 0.0042463080026209354, 'train/grad_norm_pre_clip': 2.208815574645996, 'timing/data': 0.0001932741142809391, 'timing/model': 1.2320565960835665, 'learning_rate/action_model': 8.236336777182998e-05, 'learning_rate/action_query_proj': 8.236336777182998e-05, 'learning_rate/base': 2.0590841942957496e-05, 'epoch': 0.24}) +[2026-06-29 20:38:25,651][starVLA.training.train_starvla][INFO] - Step 565, Loss: {'train/loss': 0.17171616852283478, 'train/grad_norm_pre_clip': 3.5845370292663574, 'timing/data': 0.00023401109501719475, 'timing/model': 1.2345824849326164, 'learning_rate/action_model': 8.230493418157014e-05, 'learning_rate/action_query_proj': 8.230493418157014e-05, 'learning_rate/base': 2.0576233545392536e-05, 'epoch': 0.24}) +[2026-06-29 20:38:58,347][starVLA.training.train_starvla][INFO] - Step 566, Loss: {'train/loss': 0.05804802477359772, 'train/grad_norm_pre_clip': 2.483285665512085, 'timing/data': 0.00035949796438217163, 'timing/model': 1.2340021168347448, 'learning_rate/action_model': 8.224642581689773e-05, 'learning_rate/action_query_proj': 8.224642581689773e-05, 'learning_rate/base': 2.0561606454224433e-05, 'epoch': 0.24}) +[2026-06-29 20:39:31,070][starVLA.training.train_starvla][INFO] - Step 567, Loss: {'train/loss': 0.028588756918907166, 'train/grad_norm_pre_clip': 2.1094534397125244, 'timing/data': 0.0003570341505110264, 'timing/model': 1.2313556338194758, 'learning_rate/action_model': 8.218784282217633e-05, 'learning_rate/action_query_proj': 8.218784282217633e-05, 'learning_rate/base': 2.054696070554408e-05, 'epoch': 0.24}) +[2026-06-29 20:40:03,741][starVLA.training.train_starvla][INFO] - Step 568, Loss: {'train/loss': 0.003460169769823551, 'train/grad_norm_pre_clip': 1.801814317703247, 'timing/data': 0.00023702695034444332, 'timing/model': 1.2305559900123626, 'learning_rate/action_model': 8.212918534195363e-05, 'learning_rate/action_query_proj': 8.212918534195363e-05, 'learning_rate/base': 2.0532296335488408e-05, 'epoch': 0.24}) +[2026-06-29 20:40:36,432][starVLA.training.train_starvla][INFO] - Step 569, Loss: {'train/loss': 0.009305273182690144, 'train/grad_norm_pre_clip': 2.9976913928985596, 'timing/data': 0.0003817998804152012, 'timing/model': 1.2302204947918653, 'learning_rate/action_model': 8.207045352096116e-05, 'learning_rate/action_query_proj': 8.207045352096116e-05, 'learning_rate/base': 2.051761338024029e-05, 'epoch': 0.24}) +[2026-06-29 20:41:09,147][starVLA.training.train_starvla][INFO] - Step 570, Loss: {'train/loss': 0.006192804779857397, 'train/grad_norm_pre_clip': 1.5033214092254639, 'timing/data': 0.0003023301251232624, 'timing/model': 1.2338163920212537, 'learning_rate/action_model': 8.201164750411387e-05, 'learning_rate/action_query_proj': 8.201164750411387e-05, 'learning_rate/base': 2.0502911876028467e-05, 'epoch': 0.25}) +[2026-06-29 20:41:41,858][starVLA.training.train_starvla][INFO] - Step 571, Loss: {'train/loss': 0.03650311008095741, 'train/grad_norm_pre_clip': 5.0733113288879395, 'timing/data': 0.0002888550516217947, 'timing/model': 1.2353715898934752, 'learning_rate/action_model': 8.19527674365097e-05, 'learning_rate/action_query_proj': 8.19527674365097e-05, 'learning_rate/base': 2.0488191859127426e-05, 'epoch': 0.25}) +[2026-06-29 20:42:14,582][starVLA.training.train_starvla][INFO] - Step 572, Loss: {'train/loss': 0.0019271986093372107, 'train/grad_norm_pre_clip': 2.329164505004883, 'timing/data': 0.0002501020208001137, 'timing/model': 1.2319390920456499, 'learning_rate/action_model': 8.18938134634294e-05, 'learning_rate/action_query_proj': 8.18938134634294e-05, 'learning_rate/base': 2.047345336585735e-05, 'epoch': 0.25}) +[2026-06-29 20:42:47,486][starVLA.training.train_starvla][INFO] - Step 573, Loss: {'train/loss': 0.08798728138208389, 'train/grad_norm_pre_clip': 3.1998207569122314, 'timing/data': 0.00020052888430655003, 'timing/model': 1.2281423981767148, 'learning_rate/action_model': 8.183478573033605e-05, 'learning_rate/action_query_proj': 8.183478573033605e-05, 'learning_rate/base': 2.0458696432584012e-05, 'epoch': 0.25}) +[2026-06-29 20:43:20,173][starVLA.training.train_starvla][INFO] - Step 574, Loss: {'train/loss': 0.004149754531681538, 'train/grad_norm_pre_clip': 2.330500364303589, 'timing/data': 0.00025643385015428066, 'timing/model': 1.233731335029006, 'learning_rate/action_model': 8.17756843828747e-05, 'learning_rate/action_query_proj': 8.17756843828747e-05, 'learning_rate/base': 2.0443921095718674e-05, 'epoch': 0.25}) +[2026-06-29 20:43:52,892][starVLA.training.train_starvla][INFO] - Step 575, Loss: {'train/loss': 0.0005238990415818989, 'train/grad_norm_pre_clip': 3.7938852310180664, 'timing/data': 0.0001923530362546444, 'timing/model': 1.2392410549800843, 'learning_rate/action_model': 8.171650956687204e-05, 'learning_rate/action_query_proj': 8.171650956687204e-05, 'learning_rate/base': 2.042912739171801e-05, 'epoch': 0.25}) +[2026-06-29 20:44:25,589][starVLA.training.train_starvla][INFO] - Step 576, Loss: {'train/loss': 0.44999974966049194, 'train/grad_norm_pre_clip': 8.512445449829102, 'timing/data': 0.0002561339642852545, 'timing/model': 1.2308548798318952, 'learning_rate/action_model': 8.165726142833606e-05, 'learning_rate/action_query_proj': 8.165726142833606e-05, 'learning_rate/base': 2.0414315357084015e-05, 'epoch': 0.25}) +[2026-06-29 20:44:58,251][starVLA.training.train_starvla][INFO] - Step 577, Loss: {'train/loss': 0.022932693362236023, 'train/grad_norm_pre_clip': 2.8267157077789307, 'timing/data': 0.000246644951403141, 'timing/model': 1.2354613898787647, 'learning_rate/action_model': 8.159794011345564e-05, 'learning_rate/action_query_proj': 8.159794011345564e-05, 'learning_rate/base': 2.039948502836391e-05, 'epoch': 0.25}) +[2026-06-29 20:45:30,919][starVLA.training.train_starvla][INFO] - Step 578, Loss: {'train/loss': 0.03930574283003807, 'train/grad_norm_pre_clip': 5.388875484466553, 'timing/data': 0.0001969409640878439, 'timing/model': 1.2285307450219989, 'learning_rate/action_model': 8.153854576860024e-05, 'learning_rate/action_query_proj': 8.153854576860024e-05, 'learning_rate/base': 2.038463644215006e-05, 'epoch': 0.25}) +[2026-06-29 20:46:03,570][starVLA.training.train_starvla][INFO] - Step 579, Loss: {'train/loss': 0.09983082860708237, 'train/grad_norm_pre_clip': 3.492631673812866, 'timing/data': 0.0002720730844885111, 'timing/model': 1.23219227604568, 'learning_rate/action_model': 8.14790785403195e-05, 'learning_rate/action_query_proj': 8.14790785403195e-05, 'learning_rate/base': 2.0369769635079874e-05, 'epoch': 0.25}) +[2026-06-29 20:46:36,318][starVLA.training.train_starvla][INFO] - Step 580, Loss: {'train/loss': 0.3456902801990509, 'train/grad_norm_pre_clip': 2.2121291160583496, 'timing/data': 0.0002776139881461859, 'timing/model': 1.2288936071563512, 'learning_rate/action_model': 8.141953857534288e-05, 'learning_rate/action_query_proj': 8.141953857534288e-05, 'learning_rate/base': 2.035488464383572e-05, 'epoch': 0.25}) +[2026-06-29 20:47:08,977][starVLA.training.train_starvla][INFO] - Step 581, Loss: {'train/loss': 0.006292133592069149, 'train/grad_norm_pre_clip': 2.151841163635254, 'timing/data': 0.00025569298304617405, 'timing/model': 1.2267442180309445, 'learning_rate/action_model': 8.135992602057934e-05, 'learning_rate/action_query_proj': 8.135992602057934e-05, 'learning_rate/base': 2.0339981505144835e-05, 'epoch': 0.25}) +[2026-06-29 20:47:41,634][starVLA.training.train_starvla][INFO] - Step 582, Loss: {'train/loss': 0.0006048532668501139, 'train/grad_norm_pre_clip': 2.4071974754333496, 'timing/data': 0.00028538797050714493, 'timing/model': 1.2280929600819945, 'learning_rate/action_model': 8.130024102311692e-05, 'learning_rate/action_query_proj': 8.130024102311692e-05, 'learning_rate/base': 2.032506025577923e-05, 'epoch': 0.25}) +[2026-06-29 20:48:14,354][starVLA.training.train_starvla][INFO] - Step 583, Loss: {'train/loss': 0.053072888404130936, 'train/grad_norm_pre_clip': 1.7017621994018555, 'timing/data': 0.00018966780044138432, 'timing/model': 1.2324968068860471, 'learning_rate/action_model': 8.124048373022244e-05, 'learning_rate/action_query_proj': 8.124048373022244e-05, 'learning_rate/base': 2.031012093255561e-05, 'epoch': 0.25}) +[2026-06-29 20:48:47,061][starVLA.training.train_starvla][INFO] - Step 584, Loss: {'train/loss': 0.09059184044599533, 'train/grad_norm_pre_clip': 2.247908115386963, 'timing/data': 0.0002063601277768612, 'timing/model': 1.229653764050454, 'learning_rate/action_model': 8.118065428934107e-05, 'learning_rate/action_query_proj': 8.118065428934107e-05, 'learning_rate/base': 2.0295163572335267e-05, 'epoch': 0.25}) +[2026-06-29 20:49:19,770][starVLA.training.train_starvla][INFO] - Step 585, Loss: {'train/loss': 0.030344905331730843, 'train/grad_norm_pre_clip': 4.9693217277526855, 'timing/data': 0.00019984692335128784, 'timing/model': 1.2336870639119297, 'learning_rate/action_model': 8.112075284809598e-05, 'learning_rate/action_query_proj': 8.112075284809598e-05, 'learning_rate/base': 2.0280188212023995e-05, 'epoch': 0.25}) +[2026-06-29 20:49:52,521][starVLA.training.train_starvla][INFO] - Step 586, Loss: {'train/loss': 8.063777204370126e-05, 'train/grad_norm_pre_clip': 2.641530990600586, 'timing/data': 0.00026782508939504623, 'timing/model': 1.236145885195583, 'learning_rate/action_model': 8.106077955428806e-05, 'learning_rate/action_query_proj': 8.106077955428806e-05, 'learning_rate/base': 2.0265194888572016e-05, 'epoch': 0.25}) +[2026-06-29 20:50:25,207][starVLA.training.train_starvla][INFO] - Step 587, Loss: {'train/loss': 0.21955251693725586, 'train/grad_norm_pre_clip': 1.4290903806686401, 'timing/data': 0.00019808392971754074, 'timing/model': 1.2310738011728972, 'learning_rate/action_model': 8.100073455589546e-05, 'learning_rate/action_query_proj': 8.100073455589546e-05, 'learning_rate/base': 2.0250183638973864e-05, 'epoch': 0.25}) +[2026-06-29 20:50:57,912][starVLA.training.train_starvla][INFO] - Step 588, Loss: {'train/loss': 0.12416204810142517, 'train/grad_norm_pre_clip': 4.688259601593018, 'timing/data': 0.0001954878680408001, 'timing/model': 1.2307506850920618, 'learning_rate/action_model': 8.09406180010732e-05, 'learning_rate/action_query_proj': 8.09406180010732e-05, 'learning_rate/base': 2.02351545002683e-05, 'epoch': 0.25}) +[2026-06-29 20:51:30,618][starVLA.training.train_starvla][INFO] - Step 589, Loss: {'train/loss': 0.338224858045578, 'train/grad_norm_pre_clip': 8.21129035949707, 'timing/data': 0.00020126882009208202, 'timing/model': 1.233242582064122, 'learning_rate/action_model': 8.088043003815293e-05, 'learning_rate/action_query_proj': 8.088043003815293e-05, 'learning_rate/base': 2.022010750953823e-05, 'epoch': 0.25}) +[2026-06-29 20:52:03,287][starVLA.training.train_starvla][INFO] - Step 590, Loss: {'train/loss': 0.0875038132071495, 'train/grad_norm_pre_clip': 4.249603748321533, 'timing/data': 0.00020111002959311008, 'timing/model': 1.2314228750765324, 'learning_rate/action_model': 8.082017081564244e-05, 'learning_rate/action_query_proj': 8.082017081564244e-05, 'learning_rate/base': 2.020504270391061e-05, 'epoch': 0.25}) +[2026-06-29 20:52:35,947][starVLA.training.train_starvla][INFO] - Step 591, Loss: {'train/loss': 0.11333957314491272, 'train/grad_norm_pre_clip': 1.3165568113327026, 'timing/data': 0.00025698402896523476, 'timing/model': 1.2287610510829836, 'learning_rate/action_model': 8.07598404822254e-05, 'learning_rate/action_query_proj': 8.07598404822254e-05, 'learning_rate/base': 2.018996012055635e-05, 'epoch': 0.25}) +[2026-06-29 20:53:08,649][starVLA.training.train_starvla][INFO] - Step 592, Loss: {'train/loss': 0.026148075237870216, 'train/grad_norm_pre_clip': 1.084174633026123, 'timing/data': 0.00025228620506823063, 'timing/model': 1.2280452591367066, 'learning_rate/action_model': 8.069943918676091e-05, 'learning_rate/action_query_proj': 8.069943918676091e-05, 'learning_rate/base': 2.017485979669023e-05, 'epoch': 0.25}) +[2026-06-29 20:53:41,358][starVLA.training.train_starvla][INFO] - Step 593, Loss: {'train/loss': 0.007539937738329172, 'train/grad_norm_pre_clip': 2.964463949203491, 'timing/data': 0.00019655097275972366, 'timing/model': 1.2319621979258955, 'learning_rate/action_model': 8.063896707828317e-05, 'learning_rate/action_query_proj': 8.063896707828317e-05, 'learning_rate/base': 2.0159741769570793e-05, 'epoch': 0.26}) +[2026-06-29 20:54:14,068][starVLA.training.train_starvla][INFO] - Step 594, Loss: {'train/loss': 0.38952019810676575, 'train/grad_norm_pre_clip': 5.729782581329346, 'timing/data': 0.00038927397690713406, 'timing/model': 1.228995620040223, 'learning_rate/action_model': 8.057842430600105e-05, 'learning_rate/action_query_proj': 8.057842430600105e-05, 'learning_rate/base': 2.0144606076500262e-05, 'epoch': 0.26}) +[2026-06-29 20:54:46,756][starVLA.training.train_starvla][INFO] - Step 595, Loss: {'train/loss': 0.019911307841539383, 'train/grad_norm_pre_clip': 1.77593994140625, 'timing/data': 0.0003222378436475992, 'timing/model': 1.2317660369444638, 'learning_rate/action_model': 8.051781101929786e-05, 'learning_rate/action_query_proj': 8.051781101929786e-05, 'learning_rate/base': 2.0129452754824466e-05, 'epoch': 0.26}) +[2026-06-29 20:55:19,412][starVLA.training.train_starvla][INFO] - Step 596, Loss: {'train/loss': 0.13368211686611176, 'train/grad_norm_pre_clip': 3.953712224960327, 'timing/data': 0.0003231600858271122, 'timing/model': 1.2299396628513932, 'learning_rate/action_model': 8.045712736773086e-05, 'learning_rate/action_query_proj': 8.045712736773086e-05, 'learning_rate/base': 2.0114281841932714e-05, 'epoch': 0.26}) +[2026-06-29 20:55:52,094][starVLA.training.train_starvla][INFO] - Step 597, Loss: {'train/loss': 0.01593526266515255, 'train/grad_norm_pre_clip': 1.919785976409912, 'timing/data': 0.0002724439837038517, 'timing/model': 1.2293285510968417, 'learning_rate/action_model': 8.039637350103089e-05, 'learning_rate/action_query_proj': 8.039637350103089e-05, 'learning_rate/base': 2.0099093375257723e-05, 'epoch': 0.26}) +[2026-06-29 20:56:24,783][starVLA.training.train_starvla][INFO] - Step 598, Loss: {'train/loss': 0.04536919295787811, 'train/grad_norm_pre_clip': 4.482578754425049, 'timing/data': 0.00032922113314270973, 'timing/model': 1.2294536489062011, 'learning_rate/action_model': 8.033554956910214e-05, 'learning_rate/action_query_proj': 8.033554956910214e-05, 'learning_rate/base': 2.0083887392275534e-05, 'epoch': 0.26}) +[2026-06-29 20:56:57,444][starVLA.training.train_starvla][INFO] - Step 599, Loss: {'train/loss': 0.11247977614402771, 'train/grad_norm_pre_clip': 2.4222540855407715, 'timing/data': 0.00035950797609984875, 'timing/model': 1.2308438040781766, 'learning_rate/action_model': 8.027465572202159e-05, 'learning_rate/action_query_proj': 8.027465572202159e-05, 'learning_rate/base': 2.0068663930505397e-05, 'epoch': 0.26}) +[2026-06-29 20:57:30,094][starVLA.training.train_starvla][INFO] - Step 600, Loss: {'train/loss': 0.016127776354551315, 'train/grad_norm_pre_clip': 3.4906504154205322, 'timing/data': 0.00047745020128786564, 'timing/model': 1.2310286450665444, 'learning_rate/action_model': 8.021369211003871e-05, 'learning_rate/action_query_proj': 8.021369211003871e-05, 'learning_rate/base': 2.0053423027509678e-05, 'epoch': 0.26}) +[2026-06-29 20:58:02,766][starVLA.training.train_starvla][INFO] - Step 601, Loss: {'train/loss': 0.04302248731255531, 'train/grad_norm_pre_clip': 3.4350321292877197, 'timing/data': 0.00019813398830592632, 'timing/model': 1.2281373648438603, 'learning_rate/action_model': 8.015265888357521e-05, 'learning_rate/action_query_proj': 8.015265888357521e-05, 'learning_rate/base': 2.0038164720893802e-05, 'epoch': 0.26}) +[2026-06-29 20:58:35,422][starVLA.training.train_starvla][INFO] - Step 602, Loss: {'train/loss': 2.9085969799780287e-05, 'train/grad_norm_pre_clip': 4.926265239715576, 'timing/data': 0.0001966410782188177, 'timing/model': 1.2298599381465465, 'learning_rate/action_model': 8.00915561932245e-05, 'learning_rate/action_query_proj': 8.00915561932245e-05, 'learning_rate/base': 2.0022889048306124e-05, 'epoch': 0.26}) +[2026-06-29 20:59:08,125][starVLA.training.train_starvla][INFO] - Step 603, Loss: {'train/loss': 0.13328298926353455, 'train/grad_norm_pre_clip': 2.2106053829193115, 'timing/data': 0.00019814493134617805, 'timing/model': 1.231517223874107, 'learning_rate/action_model': 8.003038418975136e-05, 'learning_rate/action_query_proj': 8.003038418975136e-05, 'learning_rate/base': 2.000759604743784e-05, 'epoch': 0.26}) +[2026-06-29 20:59:40,839][starVLA.training.train_starvla][INFO] - Step 604, Loss: {'train/loss': 0.06603674590587616, 'train/grad_norm_pre_clip': 3.115406036376953, 'timing/data': 0.00019184197299182415, 'timing/model': 1.2343243330251426, 'learning_rate/action_model': 7.996914302409167e-05, 'learning_rate/action_query_proj': 7.996914302409167e-05, 'learning_rate/base': 1.9992285756022917e-05, 'epoch': 0.26}) +[2026-06-29 21:00:13,512][starVLA.training.train_starvla][INFO] - Step 605, Loss: {'train/loss': 0.0015692973975092173, 'train/grad_norm_pre_clip': 2.231553316116333, 'timing/data': 0.00020503695122897625, 'timing/model': 1.233137388015166, 'learning_rate/action_model': 7.990783284735188e-05, 'learning_rate/action_query_proj': 7.990783284735188e-05, 'learning_rate/base': 1.997695821183797e-05, 'epoch': 0.26}) +[2026-06-29 21:00:46,189][starVLA.training.train_starvla][INFO] - Step 606, Loss: {'train/loss': 0.1447628289461136, 'train/grad_norm_pre_clip': 1.7882726192474365, 'timing/data': 0.00029486697167158127, 'timing/model': 1.2282340058591217, 'learning_rate/action_model': 7.984645381080878e-05, 'learning_rate/action_query_proj': 7.984645381080878e-05, 'learning_rate/base': 1.9961613452702194e-05, 'epoch': 0.26}) +[2026-06-29 21:01:18,884][starVLA.training.train_starvla][INFO] - Step 607, Loss: {'train/loss': 0.09984929114580154, 'train/grad_norm_pre_clip': 2.975144624710083, 'timing/data': 0.0002655519638210535, 'timing/model': 1.2351284169126302, 'learning_rate/action_model': 7.978500606590903e-05, 'learning_rate/action_query_proj': 7.978500606590903e-05, 'learning_rate/base': 1.9946251516477257e-05, 'epoch': 0.26}) +[2026-06-29 21:01:51,570][starVLA.training.train_starvla][INFO] - Step 608, Loss: {'train/loss': 0.019693318754434586, 'train/grad_norm_pre_clip': 3.753141403198242, 'timing/data': 0.000192894134670496, 'timing/model': 1.2328466170001775, 'learning_rate/action_model': 7.972348976426885e-05, 'learning_rate/action_query_proj': 7.972348976426885e-05, 'learning_rate/base': 1.9930872441067212e-05, 'epoch': 0.26}) +[2026-06-29 21:02:24,296][starVLA.training.train_starvla][INFO] - Step 609, Loss: {'train/loss': 0.027480004355311394, 'train/grad_norm_pre_clip': 3.3980674743652344, 'timing/data': 0.00019365595653653145, 'timing/model': 1.2263708170503378, 'learning_rate/action_model': 7.966190505767358e-05, 'learning_rate/action_query_proj': 7.966190505767358e-05, 'learning_rate/base': 1.9915476264418394e-05, 'epoch': 0.26}) +[2026-06-29 21:02:57,024][starVLA.training.train_starvla][INFO] - Step 610, Loss: {'train/loss': 0.0472286231815815, 'train/grad_norm_pre_clip': 3.1265180110931396, 'timing/data': 0.00029780203476548195, 'timing/model': 1.231254908023402, 'learning_rate/action_model': 7.960025209807738e-05, 'learning_rate/action_query_proj': 7.960025209807738e-05, 'learning_rate/base': 1.9900063024519346e-05, 'epoch': 0.26}) +[2026-06-29 21:03:29,699][starVLA.training.train_starvla][INFO] - Step 611, Loss: {'train/loss': 0.00204540160484612, 'train/grad_norm_pre_clip': 2.994385242462158, 'timing/data': 0.00020238198339939117, 'timing/model': 1.2320929749403149, 'learning_rate/action_model': 7.953853103760278e-05, 'learning_rate/action_query_proj': 7.953853103760278e-05, 'learning_rate/base': 1.9884632759400694e-05, 'epoch': 0.26}) +[2026-06-29 21:04:02,422][starVLA.training.train_starvla][INFO] - Step 612, Loss: {'train/loss': 0.024105586111545563, 'train/grad_norm_pre_clip': 2.4895899295806885, 'timing/data': 0.00019961711950600147, 'timing/model': 1.2341876230202615, 'learning_rate/action_model': 7.947674202854037e-05, 'learning_rate/action_query_proj': 7.947674202854037e-05, 'learning_rate/base': 1.9869185507135094e-05, 'epoch': 0.26}) +[2026-06-29 21:04:35,133][starVLA.training.train_starvla][INFO] - Step 613, Loss: {'train/loss': 0.34698060154914856, 'train/grad_norm_pre_clip': 1.6040985584259033, 'timing/data': 0.00030260090716183186, 'timing/model': 1.229223650880158, 'learning_rate/action_model': 7.941488522334838e-05, 'learning_rate/action_query_proj': 7.941488522334838e-05, 'learning_rate/base': 1.9853721305837095e-05, 'epoch': 0.26}) +[2026-06-29 21:05:07,843][starVLA.training.train_starvla][INFO] - Step 614, Loss: {'train/loss': 0.03414187580347061, 'train/grad_norm_pre_clip': 3.3573946952819824, 'timing/data': 0.00018909689970314503, 'timing/model': 1.2303356390912086, 'learning_rate/action_model': 7.935296077465235e-05, 'learning_rate/action_query_proj': 7.935296077465235e-05, 'learning_rate/base': 1.9838240193663088e-05, 'epoch': 0.26}) +[2026-06-29 21:05:40,533][starVLA.training.train_starvla][INFO] - Step 615, Loss: {'train/loss': 0.0092058926820755, 'train/grad_norm_pre_clip': 1.7181987762451172, 'timing/data': 0.00020339409820735455, 'timing/model': 1.230864271055907, 'learning_rate/action_model': 7.929096883524468e-05, 'learning_rate/action_query_proj': 7.929096883524468e-05, 'learning_rate/base': 1.982274220881117e-05, 'epoch': 0.26}) +[2026-06-29 21:06:13,226][starVLA.training.train_starvla][INFO] - Step 616, Loss: {'train/loss': 0.0037559980992227793, 'train/grad_norm_pre_clip': 5.29940938949585, 'timing/data': 0.00020031793974339962, 'timing/model': 1.230282201897353, 'learning_rate/action_model': 7.922890955808432e-05, 'learning_rate/action_query_proj': 7.922890955808432e-05, 'learning_rate/base': 1.980722738952108e-05, 'epoch': 0.27}) +[2026-06-29 21:06:45,923][starVLA.training.train_starvla][INFO] - Step 617, Loss: {'train/loss': 0.03144720941781998, 'train/grad_norm_pre_clip': 1.536399006843567, 'timing/data': 0.00029317382723093033, 'timing/model': 1.2297339518554509, 'learning_rate/action_model': 7.916678309629637e-05, 'learning_rate/action_query_proj': 7.916678309629637e-05, 'learning_rate/base': 1.979169577407409e-05, 'epoch': 0.27}) +[2026-06-29 21:07:18,607][starVLA.training.train_starvla][INFO] - Step 618, Loss: {'train/loss': 0.0252402164041996, 'train/grad_norm_pre_clip': 2.148998260498047, 'timing/data': 0.0002926930319517851, 'timing/model': 1.231540613109246, 'learning_rate/action_model': 7.910458960317171e-05, 'learning_rate/action_query_proj': 7.910458960317171e-05, 'learning_rate/base': 1.9776147400792928e-05, 'epoch': 0.27}) +[2026-06-29 21:07:51,275][starVLA.training.train_starvla][INFO] - Step 619, Loss: {'train/loss': 0.05146436020731926, 'train/grad_norm_pre_clip': 2.048089027404785, 'timing/data': 0.00025382894091308117, 'timing/model': 1.2276603851933032, 'learning_rate/action_model': 7.904232923216657e-05, 'learning_rate/action_query_proj': 7.904232923216657e-05, 'learning_rate/base': 1.9760582308041643e-05, 'epoch': 0.27}) +[2026-06-29 21:08:23,934][starVLA.training.train_starvla][INFO] - Step 620, Loss: {'train/loss': 0.08016232401132584, 'train/grad_norm_pre_clip': 1.5176759958267212, 'timing/data': 0.00023344089277088642, 'timing/model': 1.2306847339496017, 'learning_rate/action_model': 7.898000213690227e-05, 'learning_rate/action_query_proj': 7.898000213690227e-05, 'learning_rate/base': 1.974500053422557e-05, 'epoch': 0.27}) +[2026-06-29 21:08:56,675][starVLA.training.train_starvla][INFO] - Step 621, Loss: {'train/loss': 0.05263129621744156, 'train/grad_norm_pre_clip': 1.104494333267212, 'timing/data': 0.00043555302545428276, 'timing/model': 1.2337217300664634, 'learning_rate/action_model': 7.891760847116469e-05, 'learning_rate/action_query_proj': 7.891760847116469e-05, 'learning_rate/base': 1.9729402117791173e-05, 'epoch': 0.27}) +[2026-06-29 21:09:29,342][starVLA.training.train_starvla][INFO] - Step 622, Loss: {'train/loss': 0.13316355645656586, 'train/grad_norm_pre_clip': 2.6925954818725586, 'timing/data': 0.00020930496975779533, 'timing/model': 1.2331289700232446, 'learning_rate/action_model': 7.885514838890402e-05, 'learning_rate/action_query_proj': 7.885514838890402e-05, 'learning_rate/base': 1.9713787097226005e-05, 'epoch': 0.27}) +[2026-06-29 21:10:02,048][starVLA.training.train_starvla][INFO] - Step 623, Loss: {'train/loss': 0.41497451066970825, 'train/grad_norm_pre_clip': 2.7734711170196533, 'timing/data': 0.00040029617957770824, 'timing/model': 1.2318098719697446, 'learning_rate/action_model': 7.879262204423429e-05, 'learning_rate/action_query_proj': 7.879262204423429e-05, 'learning_rate/base': 1.9698155511058573e-05, 'epoch': 0.27}) +[2026-06-29 21:10:34,656][starVLA.training.train_starvla][INFO] - Step 624, Loss: {'train/loss': 0.012988989241421223, 'train/grad_norm_pre_clip': 3.684823513031006, 'timing/data': 0.00019949697889387608, 'timing/model': 1.231039104051888, 'learning_rate/action_model': 7.873002959143304e-05, 'learning_rate/action_query_proj': 7.873002959143304e-05, 'learning_rate/base': 1.968250739785826e-05, 'epoch': 0.27}) +[2026-06-29 21:11:07,358][starVLA.training.train_starvla][INFO] - Step 625, Loss: {'train/loss': 0.06413940340280533, 'train/grad_norm_pre_clip': 1.853423833847046, 'timing/data': 0.00026092189364135265, 'timing/model': 1.2300927871838212, 'learning_rate/action_model': 7.866737118494091e-05, 'learning_rate/action_query_proj': 7.866737118494091e-05, 'learning_rate/base': 1.9666842796235227e-05, 'epoch': 0.27}) +[2026-06-29 21:11:40,281][starVLA.training.train_starvla][INFO] - Step 626, Loss: {'train/loss': 0.0012432251824066043, 'train/grad_norm_pre_clip': 2.766007900238037, 'timing/data': 0.00020173098891973495, 'timing/model': 1.235478684073314, 'learning_rate/action_model': 7.860464697936132e-05, 'learning_rate/action_query_proj': 7.860464697936132e-05, 'learning_rate/base': 1.965116174484033e-05, 'epoch': 0.27}) +[2026-06-29 21:12:12,965][starVLA.training.train_starvla][INFO] - Step 627, Loss: {'train/loss': 0.0022273235954344273, 'train/grad_norm_pre_clip': 1.4786286354064941, 'timing/data': 0.00019055907614529133, 'timing/model': 1.2324156190734357, 'learning_rate/action_model': 7.854185712946001e-05, 'learning_rate/action_query_proj': 7.854185712946001e-05, 'learning_rate/base': 1.9635464282365003e-05, 'epoch': 0.27}) +[2026-06-29 21:12:45,677][starVLA.training.train_starvla][INFO] - Step 628, Loss: {'train/loss': 0.008358540013432503, 'train/grad_norm_pre_clip': 2.440725564956665, 'timing/data': 0.0002506030723452568, 'timing/model': 1.2381914360448718, 'learning_rate/action_model': 7.847900179016466e-05, 'learning_rate/action_query_proj': 7.847900179016466e-05, 'learning_rate/base': 1.9619750447541165e-05, 'epoch': 0.27}) +[2026-06-29 21:13:18,410][starVLA.training.train_starvla][INFO] - Step 629, Loss: {'train/loss': 0.02650964818894863, 'train/grad_norm_pre_clip': 3.8681905269622803, 'timing/data': 0.0002469460014253855, 'timing/model': 1.23577213101089, 'learning_rate/action_model': 7.84160811165646e-05, 'learning_rate/action_query_proj': 7.84160811165646e-05, 'learning_rate/base': 1.960402027914115e-05, 'epoch': 0.27}) +[2026-06-29 21:13:51,089][starVLA.training.train_starvla][INFO] - Step 630, Loss: {'train/loss': 0.003304707817733288, 'train/grad_norm_pre_clip': 4.339025020599365, 'timing/data': 0.0002473869826644659, 'timing/model': 1.2315010949969292, 'learning_rate/action_model': 7.835309526391033e-05, 'learning_rate/action_query_proj': 7.835309526391033e-05, 'learning_rate/base': 1.958827381597758e-05, 'epoch': 0.27}) +[2026-06-29 21:14:23,802][starVLA.training.train_starvla][INFO] - Step 631, Loss: {'train/loss': 0.00027056055841967463, 'train/grad_norm_pre_clip': 2.015714645385742, 'timing/data': 0.0002071908675134182, 'timing/model': 1.228316771099344, 'learning_rate/action_model': 7.829004438761318e-05, 'learning_rate/action_query_proj': 7.829004438761318e-05, 'learning_rate/base': 1.9572511096903296e-05, 'epoch': 0.27}) +[2026-06-29 21:14:56,519][starVLA.training.train_starvla][INFO] - Step 632, Loss: {'train/loss': 0.012972681783139706, 'train/grad_norm_pre_clip': 1.5125802755355835, 'timing/data': 0.0003164170775562525, 'timing/model': 1.2290846081450582, 'learning_rate/action_model': 7.822692864324492e-05, 'learning_rate/action_query_proj': 7.822692864324492e-05, 'learning_rate/base': 1.955673216081123e-05, 'epoch': 0.27}) +[2026-06-29 21:15:29,240][starVLA.training.train_starvla][INFO] - Step 633, Loss: {'train/loss': 0.0110972560942173, 'train/grad_norm_pre_clip': 2.5725290775299072, 'timing/data': 0.00025890907272696495, 'timing/model': 1.2346166160423309, 'learning_rate/action_model': 7.816374818653737e-05, 'learning_rate/action_query_proj': 7.816374818653737e-05, 'learning_rate/base': 1.9540937046634343e-05, 'epoch': 0.27}) +[2026-06-29 21:16:01,915][starVLA.training.train_starvla][INFO] - Step 634, Loss: {'train/loss': 0.4056490957736969, 'train/grad_norm_pre_clip': 1.8726996183395386, 'timing/data': 0.00037555908784270287, 'timing/model': 1.229056284064427, 'learning_rate/action_model': 7.810050317338204e-05, 'learning_rate/action_query_proj': 7.810050317338204e-05, 'learning_rate/base': 1.952512579334551e-05, 'epoch': 0.27}) +[2026-06-29 21:16:34,622][starVLA.training.train_starvla][INFO] - Step 635, Loss: {'train/loss': 0.11308492720127106, 'train/grad_norm_pre_clip': 1.8055267333984375, 'timing/data': 0.0002168689388781786, 'timing/model': 1.2341444389894605, 'learning_rate/action_model': 7.803719375982971e-05, 'learning_rate/action_query_proj': 7.803719375982971e-05, 'learning_rate/base': 1.9509298439957428e-05, 'epoch': 0.27}) +[2026-06-29 21:17:07,308][starVLA.training.train_starvla][INFO] - Step 636, Loss: {'train/loss': 0.05987822264432907, 'train/grad_norm_pre_clip': 2.2714192867279053, 'timing/data': 0.0002017510123550892, 'timing/model': 1.2297475980594754, 'learning_rate/action_model': 7.797382010209005e-05, 'learning_rate/action_query_proj': 7.797382010209005e-05, 'learning_rate/base': 1.949345502552251e-05, 'epoch': 0.27}) +[2026-06-29 21:17:39,980][starVLA.training.train_starvla][INFO] - Step 637, Loss: {'train/loss': 0.7872994542121887, 'train/grad_norm_pre_clip': 3.2799763679504395, 'timing/data': 0.00024405005387961864, 'timing/model': 1.2299287160858512, 'learning_rate/action_model': 7.791038235653126e-05, 'learning_rate/action_query_proj': 7.791038235653126e-05, 'learning_rate/base': 1.9477595589132815e-05, 'epoch': 0.27}) +[2026-06-29 21:18:12,661][starVLA.training.train_starvla][INFO] - Step 638, Loss: {'train/loss': 0.499967485666275, 'train/grad_norm_pre_clip': 5.859603404998779, 'timing/data': 0.000251624034717679, 'timing/model': 1.2341235589701682, 'learning_rate/action_model': 7.78468806796797e-05, 'learning_rate/action_query_proj': 7.78468806796797e-05, 'learning_rate/base': 1.9461720169919924e-05, 'epoch': 0.27}) +[2026-06-29 21:18:45,348][starVLA.training.train_starvla][INFO] - Step 639, Loss: {'train/loss': 0.19918695092201233, 'train/grad_norm_pre_clip': 2.5732998847961426, 'timing/data': 0.00026805605739355087, 'timing/model': 1.2303001137915999, 'learning_rate/action_model': 7.778331522821942e-05, 'learning_rate/action_query_proj': 7.778331522821942e-05, 'learning_rate/base': 1.9445828807054855e-05, 'epoch': 0.27}) +[2026-06-29 21:19:18,050][starVLA.training.train_starvla][INFO] - Step 640, Loss: {'train/loss': 0.0008376740152016282, 'train/grad_norm_pre_clip': 4.221856594085693, 'timing/data': 0.00033776811324059963, 'timing/model': 1.2357630219776183, 'learning_rate/action_model': 7.771968615899183e-05, 'learning_rate/action_query_proj': 7.771968615899183e-05, 'learning_rate/base': 1.9429921539747957e-05, 'epoch': 0.28}) +[2026-06-29 21:19:50,768][starVLA.training.train_starvla][INFO] - Step 641, Loss: {'train/loss': 0.00033014302607625723, 'train/grad_norm_pre_clip': 3.0175745487213135, 'timing/data': 0.00019318400882184505, 'timing/model': 1.2328639039769769, 'learning_rate/action_model': 7.76559936289954e-05, 'learning_rate/action_query_proj': 7.76559936289954e-05, 'learning_rate/base': 1.941399840724885e-05, 'epoch': 0.28}) +[2026-06-29 21:20:23,486][starVLA.training.train_starvla][INFO] - Step 642, Loss: {'train/loss': 0.10007165372371674, 'train/grad_norm_pre_clip': 2.554760694503784, 'timing/data': 0.00020213192328810692, 'timing/model': 1.2371687691193074, 'learning_rate/action_model': 7.759223779538505e-05, 'learning_rate/action_query_proj': 7.759223779538505e-05, 'learning_rate/base': 1.9398059448846263e-05, 'epoch': 0.28}) +[2026-06-29 21:20:56,157][starVLA.training.train_starvla][INFO] - Step 643, Loss: {'train/loss': 0.02793097496032715, 'train/grad_norm_pre_clip': 2.167494058609009, 'timing/data': 0.00020577898249030113, 'timing/model': 1.2300544290337712, 'learning_rate/action_model': 7.7528418815472e-05, 'learning_rate/action_query_proj': 7.7528418815472e-05, 'learning_rate/base': 1.9382104703868e-05, 'epoch': 0.28}) +[2026-06-29 21:21:28,865][starVLA.training.train_starvla][INFO] - Step 644, Loss: {'train/loss': 0.11425530910491943, 'train/grad_norm_pre_clip': 3.074260950088501, 'timing/data': 0.00019945716485381126, 'timing/model': 1.2308354680426419, 'learning_rate/action_model': 7.746453684672323e-05, 'learning_rate/action_query_proj': 7.746453684672323e-05, 'learning_rate/base': 1.936613421168081e-05, 'epoch': 0.28}) +[2026-06-29 21:22:01,587][starVLA.training.train_starvla][INFO] - Step 645, Loss: {'train/loss': 0.002219224814325571, 'train/grad_norm_pre_clip': 3.2024800777435303, 'timing/data': 0.00019889604300260544, 'timing/model': 1.2307650509756058, 'learning_rate/action_model': 7.740059204676116e-05, 'learning_rate/action_query_proj': 7.740059204676116e-05, 'learning_rate/base': 1.935014801169029e-05, 'epoch': 0.28}) +[2026-06-29 21:22:34,291][starVLA.training.train_starvla][INFO] - Step 646, Loss: {'train/loss': 0.012104639783501625, 'train/grad_norm_pre_clip': 2.2437515258789062, 'timing/data': 0.00026789610274136066, 'timing/model': 1.2342396241147071, 'learning_rate/action_model': 7.733658457336319e-05, 'learning_rate/action_query_proj': 7.733658457336319e-05, 'learning_rate/base': 1.9334146143340798e-05, 'epoch': 0.28}) +[2026-06-29 21:23:06,980][starVLA.training.train_starvla][INFO] - Step 647, Loss: {'train/loss': 0.003219058969989419, 'train/grad_norm_pre_clip': 1.3999781608581543, 'timing/data': 0.00033268798142671585, 'timing/model': 1.2326671539340168, 'learning_rate/action_model': 7.727251458446141e-05, 'learning_rate/action_query_proj': 7.727251458446141e-05, 'learning_rate/base': 1.9318128646115354e-05, 'epoch': 0.28}) +[2026-06-29 21:23:39,728][starVLA.training.train_starvla][INFO] - Step 648, Loss: {'train/loss': 0.001129669020883739, 'train/grad_norm_pre_clip': 3.7482385635375977, 'timing/data': 0.00029782182537019253, 'timing/model': 1.2362776238005608, 'learning_rate/action_model': 7.72083822381422e-05, 'learning_rate/action_query_proj': 7.72083822381422e-05, 'learning_rate/base': 1.930209555953555e-05, 'epoch': 0.28}) +[2026-06-29 21:24:12,385][starVLA.training.train_starvla][INFO] - Step 649, Loss: {'train/loss': 0.17255184054374695, 'train/grad_norm_pre_clip': 1.3385051488876343, 'timing/data': 0.00021786196157336235, 'timing/model': 1.2309489189647138, 'learning_rate/action_model': 7.714418769264568e-05, 'learning_rate/action_query_proj': 7.714418769264568e-05, 'learning_rate/base': 1.928604692316142e-05, 'epoch': 0.28}) +[2026-06-29 21:24:45,047][starVLA.training.train_starvla][INFO] - Step 650, Loss: {'train/loss': 0.46078839898109436, 'train/grad_norm_pre_clip': 1.5290805101394653, 'timing/data': 0.0003680239897221327, 'timing/model': 1.229458176996559, 'learning_rate/action_model': 7.707993110636554e-05, 'learning_rate/action_query_proj': 7.707993110636554e-05, 'learning_rate/base': 1.9269982776591384e-05, 'epoch': 0.28}) +[2026-06-29 21:25:17,740][starVLA.training.train_starvla][INFO] - Step 651, Loss: {'train/loss': 0.6018201112747192, 'train/grad_norm_pre_clip': 1.5500881671905518, 'timing/data': 0.00024473201483488083, 'timing/model': 1.2320094511378556, 'learning_rate/action_model': 7.701561263784853e-05, 'learning_rate/action_query_proj': 7.701561263784853e-05, 'learning_rate/base': 1.925390315946213e-05, 'epoch': 0.28}) +[2026-06-29 21:25:50,378][starVLA.training.train_starvla][INFO] - Step 652, Loss: {'train/loss': 0.005257258657366037, 'train/grad_norm_pre_clip': 1.7658041715621948, 'timing/data': 0.0002689971588551998, 'timing/model': 1.230324984062463, 'learning_rate/action_model': 7.695123244579405e-05, 'learning_rate/action_query_proj': 7.695123244579405e-05, 'learning_rate/base': 1.9237808111448512e-05, 'epoch': 0.28}) +[2026-06-29 21:26:23,077][starVLA.training.train_starvla][INFO] - Step 653, Loss: {'train/loss': 0.007455820683389902, 'train/grad_norm_pre_clip': 1.7573423385620117, 'timing/data': 0.00020120013505220413, 'timing/model': 1.2303337920457125, 'learning_rate/action_model': 7.688679068905386e-05, 'learning_rate/action_query_proj': 7.688679068905386e-05, 'learning_rate/base': 1.9221697672263465e-05, 'epoch': 0.28}) +[2026-06-29 21:26:55,799][starVLA.training.train_starvla][INFO] - Step 654, Loss: {'train/loss': 0.00976761244237423, 'train/grad_norm_pre_clip': 2.073484420776367, 'timing/data': 0.0001970308367162943, 'timing/model': 1.237948349909857, 'learning_rate/action_model': 7.682228752663154e-05, 'learning_rate/action_query_proj': 7.682228752663154e-05, 'learning_rate/base': 1.9205571881657884e-05, 'epoch': 0.28}) +[2026-06-29 21:27:28,484][starVLA.training.train_starvla][INFO] - Step 655, Loss: {'train/loss': 0.5127860307693481, 'train/grad_norm_pre_clip': 1.8064395189285278, 'timing/data': 0.00025066197849810123, 'timing/model': 1.2316744150593877, 'learning_rate/action_model': 7.675772311768228e-05, 'learning_rate/action_query_proj': 7.675772311768228e-05, 'learning_rate/base': 1.918943077942057e-05, 'epoch': 0.28}) +[2026-06-29 21:28:01,175][starVLA.training.train_starvla][INFO] - Step 656, Loss: {'train/loss': 0.03697090595960617, 'train/grad_norm_pre_clip': 2.5902228355407715, 'timing/data': 0.00019885599613189697, 'timing/model': 1.2353079940658063, 'learning_rate/action_model': 7.66930976215123e-05, 'learning_rate/action_query_proj': 7.66930976215123e-05, 'learning_rate/base': 1.9173274405378077e-05, 'epoch': 0.28}) +[2026-06-29 21:28:33,879][starVLA.training.train_starvla][INFO] - Step 657, Loss: {'train/loss': 0.008656982332468033, 'train/grad_norm_pre_clip': 1.8081165552139282, 'timing/data': 0.0001945469994097948, 'timing/model': 1.2321883321274072, 'learning_rate/action_model': 7.662841119757864e-05, 'learning_rate/action_query_proj': 7.662841119757864e-05, 'learning_rate/base': 1.915710279939466e-05, 'epoch': 0.28}) +[2026-06-29 21:29:06,575][starVLA.training.train_starvla][INFO] - Step 658, Loss: {'train/loss': 0.41799670457839966, 'train/grad_norm_pre_clip': 2.9714107513427734, 'timing/data': 0.00019415700808167458, 'timing/model': 1.2340722889639437, 'learning_rate/action_model': 7.656366400548857e-05, 'learning_rate/action_query_proj': 7.656366400548857e-05, 'learning_rate/base': 1.914091600137214e-05, 'epoch': 0.28}) +[2026-06-29 21:29:39,271][starVLA.training.train_starvla][INFO] - Step 659, Loss: {'train/loss': 0.00043224403634667397, 'train/grad_norm_pre_clip': 2.494156837463379, 'timing/data': 0.0002014001365751028, 'timing/model': 1.234719174914062, 'learning_rate/action_model': 7.649885620499935e-05, 'learning_rate/action_query_proj': 7.649885620499935e-05, 'learning_rate/base': 1.912471405124984e-05, 'epoch': 0.28}) +[2026-06-29 21:30:12,023][starVLA.training.train_starvla][INFO] - Step 660, Loss: {'train/loss': 0.006303803995251656, 'train/grad_norm_pre_clip': 1.380674958229065, 'timing/data': 0.0004116259515285492, 'timing/model': 1.2283919618930668, 'learning_rate/action_model': 7.643398795601782e-05, 'learning_rate/action_query_proj': 7.643398795601782e-05, 'learning_rate/base': 1.9108496989004455e-05, 'epoch': 0.28}) +[2026-06-29 21:30:44,706][starVLA.training.train_starvla][INFO] - Step 661, Loss: {'train/loss': 0.00017973546346183866, 'train/grad_norm_pre_clip': 1.6597528457641602, 'timing/data': 0.00025232601910829544, 'timing/model': 1.2280343321617693, 'learning_rate/action_model': 7.636905941859994e-05, 'learning_rate/action_query_proj': 7.636905941859994e-05, 'learning_rate/base': 1.9092264854649984e-05, 'epoch': 0.28}) +[2026-06-29 21:31:17,390][starVLA.training.train_starvla][INFO] - Step 662, Loss: {'train/loss': 0.06473138928413391, 'train/grad_norm_pre_clip': 2.2871053218841553, 'timing/data': 0.0002495211083441973, 'timing/model': 1.2276312271133065, 'learning_rate/action_model': 7.630407075295037e-05, 'learning_rate/action_query_proj': 7.630407075295037e-05, 'learning_rate/base': 1.907601768823759e-05, 'epoch': 0.28}) +[2026-06-29 21:31:50,092][starVLA.training.train_starvla][INFO] - Step 663, Loss: {'train/loss': 0.003391180420294404, 'train/grad_norm_pre_clip': 2.744135856628418, 'timing/data': 0.0002487790770828724, 'timing/model': 1.235051502007991, 'learning_rate/action_model': 7.623902211942222e-05, 'learning_rate/action_query_proj': 7.623902211942222e-05, 'learning_rate/base': 1.9059755529855555e-05, 'epoch': 0.29}) +[2026-06-29 21:32:22,790][starVLA.training.train_starvla][INFO] - Step 664, Loss: {'train/loss': 0.0012926501221954823, 'train/grad_norm_pre_clip': 1.6347028017044067, 'timing/data': 0.00020598899573087692, 'timing/model': 1.2286491489503533, 'learning_rate/action_model': 7.617391367851653e-05, 'learning_rate/action_query_proj': 7.617391367851653e-05, 'learning_rate/base': 1.9043478419629132e-05, 'epoch': 0.29}) +[2026-06-29 21:32:55,445][starVLA.training.train_starvla][INFO] - Step 665, Loss: {'train/loss': 0.5649445652961731, 'train/grad_norm_pre_clip': 3.1919360160827637, 'timing/data': 0.00019092997536063194, 'timing/model': 1.2301251059398055, 'learning_rate/action_model': 7.610874559088188e-05, 'learning_rate/action_query_proj': 7.610874559088188e-05, 'learning_rate/base': 1.902718639772047e-05, 'epoch': 0.29}) +[2026-06-29 21:33:28,134][starVLA.training.train_starvla][INFO] - Step 666, Loss: {'train/loss': 0.04321054741740227, 'train/grad_norm_pre_clip': 2.4116594791412354, 'timing/data': 0.00020208186469972134, 'timing/model': 1.234577322145924, 'learning_rate/action_model': 7.604351801731409e-05, 'learning_rate/action_query_proj': 7.604351801731409e-05, 'learning_rate/base': 1.9010879504328523e-05, 'epoch': 0.29}) +[2026-06-29 21:34:00,852][starVLA.training.train_starvla][INFO] - Step 667, Loss: {'train/loss': 0.009951315820217133, 'train/grad_norm_pre_clip': 2.4921209812164307, 'timing/data': 0.00019784295000135899, 'timing/model': 1.2319942510221153, 'learning_rate/action_model': 7.597823111875565e-05, 'learning_rate/action_query_proj': 7.597823111875565e-05, 'learning_rate/base': 1.8994557779688914e-05, 'epoch': 0.29}) +[2026-06-29 21:34:33,535][starVLA.training.train_starvla][INFO] - Step 668, Loss: {'train/loss': 0.9332984685897827, 'train/grad_norm_pre_clip': 7.614260673522949, 'timing/data': 0.0002760309725999832, 'timing/model': 1.2284824510570616, 'learning_rate/action_model': 7.591288505629556e-05, 'learning_rate/action_query_proj': 7.591288505629556e-05, 'learning_rate/base': 1.897822126407389e-05, 'epoch': 0.29}) +[2026-06-29 21:35:06,230][starVLA.training.train_starvla][INFO] - Step 669, Loss: {'train/loss': 0.04406850412487984, 'train/grad_norm_pre_clip': 5.57143497467041, 'timing/data': 0.00020195101387798786, 'timing/model': 1.2319825710728765, 'learning_rate/action_model': 7.58474799911687e-05, 'learning_rate/action_query_proj': 7.58474799911687e-05, 'learning_rate/base': 1.8961869997792175e-05, 'epoch': 0.29}) +[2026-06-29 21:35:38,906][starVLA.training.train_starvla][INFO] - Step 670, Loss: {'train/loss': 0.00964162964373827, 'train/grad_norm_pre_clip': 5.758005619049072, 'timing/data': 0.00032749795354902744, 'timing/model': 1.226059417007491, 'learning_rate/action_model': 7.578201608475557e-05, 'learning_rate/action_query_proj': 7.578201608475557e-05, 'learning_rate/base': 1.8945504021188893e-05, 'epoch': 0.29}) +[2026-06-29 21:36:11,605][starVLA.training.train_starvla][INFO] - Step 671, Loss: {'train/loss': 0.274025022983551, 'train/grad_norm_pre_clip': 2.8579492568969727, 'timing/data': 0.00019606994464993477, 'timing/model': 1.230395320802927, 'learning_rate/action_model': 7.571649349858185e-05, 'learning_rate/action_query_proj': 7.571649349858185e-05, 'learning_rate/base': 1.8929123374645462e-05, 'epoch': 0.29}) +[2026-06-29 21:36:44,317][starVLA.training.train_starvla][INFO] - Step 672, Loss: {'train/loss': 0.018234731629490852, 'train/grad_norm_pre_clip': 1.2136512994766235, 'timing/data': 0.0001937849447131157, 'timing/model': 1.2319487761706114, 'learning_rate/action_model': 7.5650912394318e-05, 'learning_rate/action_query_proj': 7.5650912394318e-05, 'learning_rate/base': 1.89127280985795e-05, 'epoch': 0.29}) +[2026-06-29 21:37:17,011][starVLA.training.train_starvla][INFO] - Step 673, Loss: {'train/loss': 0.023865997791290283, 'train/grad_norm_pre_clip': 4.8272480964660645, 'timing/data': 0.00022282102145254612, 'timing/model': 1.2291605600621551, 'learning_rate/action_model': 7.558527293377889e-05, 'learning_rate/action_query_proj': 7.558527293377889e-05, 'learning_rate/base': 1.8896318233444723e-05, 'epoch': 0.29}) +[2026-06-29 21:37:49,699][starVLA.training.train_starvla][INFO] - Step 674, Loss: {'train/loss': 0.0023329860996454954, 'train/grad_norm_pre_clip': 1.7478883266448975, 'timing/data': 0.0001935749314725399, 'timing/model': 1.2298180400393903, 'learning_rate/action_model': 7.551957527892335e-05, 'learning_rate/action_query_proj': 7.551957527892335e-05, 'learning_rate/base': 1.887989381973084e-05, 'epoch': 0.29}) +[2026-06-29 21:38:22,372][starVLA.training.train_starvla][INFO] - Step 675, Loss: {'train/loss': 0.25353384017944336, 'train/grad_norm_pre_clip': 1.5368797779083252, 'timing/data': 0.000198825029656291, 'timing/model': 1.2236014979425818, 'learning_rate/action_model': 7.545381959185384e-05, 'learning_rate/action_query_proj': 7.545381959185384e-05, 'learning_rate/base': 1.886345489796346e-05, 'epoch': 0.29}) +[2026-06-29 21:38:55,049][starVLA.training.train_starvla][INFO] - Step 676, Loss: {'train/loss': 0.020279865711927414, 'train/grad_norm_pre_clip': 2.00319766998291, 'timing/data': 0.0001992459874600172, 'timing/model': 1.2308834150899202, 'learning_rate/action_model': 7.538800603481596e-05, 'learning_rate/action_query_proj': 7.538800603481596e-05, 'learning_rate/base': 1.884700150870399e-05, 'epoch': 0.29}) +[2026-06-29 21:39:27,892][starVLA.training.train_starvla][INFO] - Step 677, Loss: {'train/loss': 0.006957768462598324, 'train/grad_norm_pre_clip': 2.6316328048706055, 'timing/data': 0.0004396899603307247, 'timing/model': 1.231462774099782, 'learning_rate/action_model': 7.532213477019812e-05, 'learning_rate/action_query_proj': 7.532213477019812e-05, 'learning_rate/base': 1.883053369254953e-05, 'epoch': 0.29}) +[2026-06-29 21:40:00,600][starVLA.training.train_starvla][INFO] - Step 678, Loss: {'train/loss': 0.0011350236600264907, 'train/grad_norm_pre_clip': 3.309830904006958, 'timing/data': 0.0001923029776662588, 'timing/model': 1.2341539398767054, 'learning_rate/action_model': 7.52562059605311e-05, 'learning_rate/action_query_proj': 7.52562059605311e-05, 'learning_rate/base': 1.8814051490132775e-05, 'epoch': 0.29}) +[2026-06-29 21:40:33,278][starVLA.training.train_starvla][INFO] - Step 679, Loss: {'train/loss': 0.00031215179478749633, 'train/grad_norm_pre_clip': 3.8966495990753174, 'timing/data': 0.0002170500811189413, 'timing/model': 1.2329455900471658, 'learning_rate/action_model': 7.519021976848773e-05, 'learning_rate/action_query_proj': 7.519021976848773e-05, 'learning_rate/base': 1.8797554942121932e-05, 'epoch': 0.29}) +[2026-06-29 21:41:05,964][starVLA.training.train_starvla][INFO] - Step 680, Loss: {'train/loss': 0.5258366465568542, 'train/grad_norm_pre_clip': 3.4497621059417725, 'timing/data': 0.0002900669351220131, 'timing/model': 1.2382066000718623, 'learning_rate/action_model': 7.512417635688234e-05, 'learning_rate/action_query_proj': 7.512417635688234e-05, 'learning_rate/base': 1.8781044089220585e-05, 'epoch': 0.29}) +[2026-06-29 21:41:38,668][starVLA.training.train_starvla][INFO] - Step 681, Loss: {'train/loss': 0.36244603991508484, 'train/grad_norm_pre_clip': 3.8952226638793945, 'timing/data': 0.00019994680769741535, 'timing/model': 1.2346451829653233, 'learning_rate/action_model': 7.50580758886705e-05, 'learning_rate/action_query_proj': 7.50580758886705e-05, 'learning_rate/base': 1.8764518972167624e-05, 'epoch': 0.29}) +[2026-06-29 21:42:11,370][starVLA.training.train_starvla][INFO] - Step 682, Loss: {'train/loss': 0.014118963852524757, 'train/grad_norm_pre_clip': 1.1774836778640747, 'timing/data': 0.00019679195247590542, 'timing/model': 1.2251610499806702, 'learning_rate/action_model': 7.499191852694854e-05, 'learning_rate/action_query_proj': 7.499191852694854e-05, 'learning_rate/base': 1.8747979631737135e-05, 'epoch': 0.29}) +[2026-06-29 21:42:43,987][starVLA.training.train_starvla][INFO] - Step 683, Loss: {'train/loss': 0.022137420251965523, 'train/grad_norm_pre_clip': 3.3971378803253174, 'timing/data': 0.00025019189342856407, 'timing/model': 1.2292568958364427, 'learning_rate/action_model': 7.492570443495316e-05, 'learning_rate/action_query_proj': 7.492570443495316e-05, 'learning_rate/base': 1.873142610873829e-05, 'epoch': 0.29}) +[2026-06-29 21:43:16,667][starVLA.training.train_starvla][INFO] - Step 684, Loss: {'train/loss': 0.011561321094632149, 'train/grad_norm_pre_clip': 3.107975721359253, 'timing/data': 0.00028550787828862667, 'timing/model': 1.2323584170080721, 'learning_rate/action_model': 7.485943377606109e-05, 'learning_rate/action_query_proj': 7.485943377606109e-05, 'learning_rate/base': 1.8714858444015272e-05, 'epoch': 0.29}) +[2026-06-29 21:43:49,358][starVLA.training.train_starvla][INFO] - Step 685, Loss: {'train/loss': 0.29183048009872437, 'train/grad_norm_pre_clip': 2.123100757598877, 'timing/data': 0.00020010792650282383, 'timing/model': 1.2314858599565923, 'learning_rate/action_model': 7.479310671378854e-05, 'learning_rate/action_query_proj': 7.479310671378854e-05, 'learning_rate/base': 1.8698276678447134e-05, 'epoch': 0.29}) +[2026-06-29 21:44:22,003][starVLA.training.train_starvla][INFO] - Step 686, Loss: {'train/loss': 0.37835556268692017, 'train/grad_norm_pre_clip': 3.8312840461730957, 'timing/data': 0.0003211759030818939, 'timing/model': 1.2274508669506758, 'learning_rate/action_model': 7.472672341179099e-05, 'learning_rate/action_query_proj': 7.472672341179099e-05, 'learning_rate/base': 1.8681680852947747e-05, 'epoch': 0.3}) +[2026-06-29 21:44:54,606][starVLA.training.train_starvla][INFO] - Step 687, Loss: {'train/loss': 0.1527174413204193, 'train/grad_norm_pre_clip': 5.219443321228027, 'timing/data': 0.000320865074172616, 'timing/model': 1.2267874479293823, 'learning_rate/action_model': 7.46602840338626e-05, 'learning_rate/action_query_proj': 7.46602840338626e-05, 'learning_rate/base': 1.866507100846565e-05, 'epoch': 0.3}) +[2026-06-29 21:45:27,315][starVLA.training.train_starvla][INFO] - Step 688, Loss: {'train/loss': 0.19421789050102234, 'train/grad_norm_pre_clip': 6.385371208190918, 'timing/data': 0.00025210599415004253, 'timing/model': 1.2336267970968038, 'learning_rate/action_model': 7.459378874393598e-05, 'learning_rate/action_query_proj': 7.459378874393598e-05, 'learning_rate/base': 1.8648447185983994e-05, 'epoch': 0.3}) +[2026-06-29 21:46:00,010][starVLA.training.train_starvla][INFO] - Step 689, Loss: {'train/loss': 0.14084787666797638, 'train/grad_norm_pre_clip': 2.7683379650115967, 'timing/data': 0.0002549809869378805, 'timing/model': 1.2299824529327452, 'learning_rate/action_model': 7.452723770608158e-05, 'learning_rate/action_query_proj': 7.452723770608158e-05, 'learning_rate/base': 1.8631809426520396e-05, 'epoch': 0.3}) +[2026-06-29 21:46:32,702][starVLA.training.train_starvla][INFO] - Step 690, Loss: {'train/loss': 0.016464898362755775, 'train/grad_norm_pre_clip': 3.4359943866729736, 'timing/data': 0.00023186695761978626, 'timing/model': 1.232436208985746, 'learning_rate/action_model': 7.446063108450753e-05, 'learning_rate/action_query_proj': 7.446063108450753e-05, 'learning_rate/base': 1.8615157771126882e-05, 'epoch': 0.3}) +[2026-06-29 21:47:05,417][starVLA.training.train_starvla][INFO] - Step 691, Loss: {'train/loss': 0.06116903945803642, 'train/grad_norm_pre_clip': 1.9966505765914917, 'timing/data': 0.00024227704852819443, 'timing/model': 1.2297699640039355, 'learning_rate/action_model': 7.439396904355902e-05, 'learning_rate/action_query_proj': 7.439396904355902e-05, 'learning_rate/base': 1.8598492260889756e-05, 'epoch': 0.3}) +[2026-06-29 21:47:38,082][starVLA.training.train_starvla][INFO] - Step 692, Loss: {'train/loss': 0.045476071536540985, 'train/grad_norm_pre_clip': 2.1216704845428467, 'timing/data': 0.00019500800408422947, 'timing/model': 1.2350403149612248, 'learning_rate/action_model': 7.4327251747718e-05, 'learning_rate/action_query_proj': 7.4327251747718e-05, 'learning_rate/base': 1.85818129369295e-05, 'epoch': 0.3}) +[2026-06-29 21:48:10,820][starVLA.training.train_starvla][INFO] - Step 693, Loss: {'train/loss': 0.19898322224617004, 'train/grad_norm_pre_clip': 2.8694915771484375, 'timing/data': 0.0001974320039153099, 'timing/model': 1.2339353689458221, 'learning_rate/action_model': 7.426047936160279e-05, 'learning_rate/action_query_proj': 7.426047936160279e-05, 'learning_rate/base': 1.8565119840400698e-05, 'epoch': 0.3}) +[2026-06-29 21:48:43,509][starVLA.training.train_starvla][INFO] - Step 694, Loss: {'train/loss': 0.02577734924852848, 'train/grad_norm_pre_clip': 1.3059109449386597, 'timing/data': 0.00021002604626119137, 'timing/model': 1.2326496511232108, 'learning_rate/action_model': 7.419365204996761e-05, 'learning_rate/action_query_proj': 7.419365204996761e-05, 'learning_rate/base': 1.8548413012491902e-05, 'epoch': 0.3}) +[2026-06-29 21:49:16,234][starVLA.training.train_starvla][INFO] - Step 695, Loss: {'train/loss': 0.36467307806015015, 'train/grad_norm_pre_clip': 2.518815755844116, 'timing/data': 0.00022970396094024181, 'timing/model': 1.232355625135824, 'learning_rate/action_model': 7.41267699777022e-05, 'learning_rate/action_query_proj': 7.41267699777022e-05, 'learning_rate/base': 1.853169249442555e-05, 'epoch': 0.3}) +[2026-06-29 21:49:48,897][starVLA.training.train_starvla][INFO] - Step 696, Loss: {'train/loss': 0.2006770670413971, 'train/grad_norm_pre_clip': 5.838474273681641, 'timing/data': 0.00020306394435465336, 'timing/model': 1.2299516580533236, 'learning_rate/action_model': 7.405983330983143e-05, 'learning_rate/action_query_proj': 7.405983330983143e-05, 'learning_rate/base': 1.851495832745786e-05, 'epoch': 0.3}) +[2026-06-29 21:50:21,580][starVLA.training.train_starvla][INFO] - Step 697, Loss: {'train/loss': 0.010619256645441055, 'train/grad_norm_pre_clip': 1.7141190767288208, 'timing/data': 0.00019280402921140194, 'timing/model': 1.2299227949697524, 'learning_rate/action_model': 7.399284221151486e-05, 'learning_rate/action_query_proj': 7.399284221151486e-05, 'learning_rate/base': 1.8498210552878716e-05, 'epoch': 0.3}) +[2026-06-29 21:50:54,226][starVLA.training.train_starvla][INFO] - Step 698, Loss: {'train/loss': 0.1079481691122055, 'train/grad_norm_pre_clip': 2.2715392112731934, 'timing/data': 0.00019533908925950527, 'timing/model': 1.2310482400935143, 'learning_rate/action_model': 7.392579684804638e-05, 'learning_rate/action_query_proj': 7.392579684804638e-05, 'learning_rate/base': 1.8481449212011594e-05, 'epoch': 0.3}) +[2026-06-29 21:51:26,902][starVLA.training.train_starvla][INFO] - Step 699, Loss: {'train/loss': 0.00989514123648405, 'train/grad_norm_pre_clip': 2.239997148513794, 'timing/data': 0.00048393290489912033, 'timing/model': 1.2318169618956745, 'learning_rate/action_model': 7.385869738485377e-05, 'learning_rate/action_query_proj': 7.385869738485377e-05, 'learning_rate/base': 1.8464674346213443e-05, 'epoch': 0.3}) +[2026-06-29 21:51:59,642][starVLA.training.train_starvla][INFO] - Step 700, Loss: {'train/loss': 0.6950734257698059, 'train/grad_norm_pre_clip': 2.5502071380615234, 'timing/data': 0.00019227294251322746, 'timing/model': 1.2340761870145798, 'learning_rate/action_model': 7.379154398749825e-05, 'learning_rate/action_query_proj': 7.379154398749825e-05, 'learning_rate/base': 1.8447885996874562e-05, 'epoch': 0.3}) +[2026-06-29 21:52:32,344][starVLA.training.train_starvla][INFO] - Step 701, Loss: {'train/loss': 0.01955270767211914, 'train/grad_norm_pre_clip': 1.3034329414367676, 'timing/data': 0.0002046460285782814, 'timing/model': 1.233454346191138, 'learning_rate/action_model': 7.372433682167419e-05, 'learning_rate/action_query_proj': 7.372433682167419e-05, 'learning_rate/base': 1.8431084205418548e-05, 'epoch': 0.3}) +[2026-06-29 21:53:05,000][starVLA.training.train_starvla][INFO] - Step 702, Loss: {'train/loss': 0.09009609371423721, 'train/grad_norm_pre_clip': 1.3799432516098022, 'timing/data': 0.0002415659837424755, 'timing/model': 1.2304573939181864, 'learning_rate/action_model': 7.365707605320857e-05, 'learning_rate/action_query_proj': 7.365707605320857e-05, 'learning_rate/base': 1.8414269013302144e-05, 'epoch': 0.3}) +[2026-06-29 21:53:37,662][starVLA.training.train_starvla][INFO] - Step 703, Loss: {'train/loss': 0.06447374075651169, 'train/grad_norm_pre_clip': 1.6295082569122314, 'timing/data': 0.00038981600664556026, 'timing/model': 1.2297696499153972, 'learning_rate/action_model': 7.358976184806066e-05, 'learning_rate/action_query_proj': 7.358976184806066e-05, 'learning_rate/base': 1.8397440462015165e-05, 'epoch': 0.3}) +[2026-06-29 21:54:10,335][starVLA.training.train_starvla][INFO] - Step 704, Loss: {'train/loss': 0.00020408749696798623, 'train/grad_norm_pre_clip': 1.6900020837783813, 'timing/data': 0.00026858714409172535, 'timing/model': 1.2313804528675973, 'learning_rate/action_model': 7.352239437232156e-05, 'learning_rate/action_query_proj': 7.352239437232156e-05, 'learning_rate/base': 1.838059859308039e-05, 'epoch': 0.3}) +[2026-06-29 21:54:42,999][starVLA.training.train_starvla][INFO] - Step 705, Loss: {'train/loss': 1.7306952476501465, 'train/grad_norm_pre_clip': 3.7027289867401123, 'timing/data': 0.00020492705516517162, 'timing/model': 1.2319676550105214, 'learning_rate/action_model': 7.345497379221382e-05, 'learning_rate/action_query_proj': 7.345497379221382e-05, 'learning_rate/base': 1.8363743448053456e-05, 'epoch': 0.3}) +[2026-06-29 21:55:15,679][starVLA.training.train_starvla][INFO] - Step 706, Loss: {'train/loss': 0.006711202673614025, 'train/grad_norm_pre_clip': 2.7330355644226074, 'timing/data': 0.00024267705157399178, 'timing/model': 1.2310158140026033, 'learning_rate/action_model': 7.338750027409104e-05, 'learning_rate/action_query_proj': 7.338750027409104e-05, 'learning_rate/base': 1.834687506852276e-05, 'epoch': 0.3}) +[2026-06-29 21:55:48,354][starVLA.training.train_starvla][INFO] - Step 707, Loss: {'train/loss': 0.026399409398436546, 'train/grad_norm_pre_clip': 1.5535638332366943, 'timing/data': 0.00022293999791145325, 'timing/model': 1.2294341139495373, 'learning_rate/action_model': 7.331997398443738e-05, 'learning_rate/action_query_proj': 7.331997398443738e-05, 'learning_rate/base': 1.8329993496109345e-05, 'epoch': 0.3}) +[2026-06-29 21:56:21,053][starVLA.training.train_starvla][INFO] - Step 708, Loss: {'train/loss': 0.00043153564911335707, 'train/grad_norm_pre_clip': 2.8713207244873047, 'timing/data': 0.00039379391819238663, 'timing/model': 1.2311224350705743, 'learning_rate/action_model': 7.325239508986729e-05, 'learning_rate/action_query_proj': 7.325239508986729e-05, 'learning_rate/base': 1.831309877246682e-05, 'epoch': 0.3}) +[2026-06-29 21:56:53,725][starVLA.training.train_starvla][INFO] - Step 709, Loss: {'train/loss': 0.07449153065681458, 'train/grad_norm_pre_clip': 2.9884753227233887, 'timing/data': 0.00033311895094811916, 'timing/model': 1.2298657598439604, 'learning_rate/action_model': 7.318476375712494e-05, 'learning_rate/action_query_proj': 7.318476375712494e-05, 'learning_rate/base': 1.8296190939281234e-05, 'epoch': 0.31}) +[2026-06-29 21:57:26,392][starVLA.training.train_starvla][INFO] - Step 710, Loss: {'train/loss': 0.1826571524143219, 'train/grad_norm_pre_clip': 3.3903090953826904, 'timing/data': 0.00020328303799033165, 'timing/model': 1.228860218077898, 'learning_rate/action_model': 7.311708015308394e-05, 'learning_rate/action_query_proj': 7.311708015308394e-05, 'learning_rate/base': 1.8279270038270986e-05, 'epoch': 0.31}) +[2026-06-29 21:57:59,066][starVLA.training.train_starvla][INFO] - Step 711, Loss: {'train/loss': 0.0023205990437418222, 'train/grad_norm_pre_clip': 1.7582942247390747, 'timing/data': 0.00020258198492228985, 'timing/model': 1.2185775830876082, 'learning_rate/action_model': 7.304934444474684e-05, 'learning_rate/action_query_proj': 7.304934444474684e-05, 'learning_rate/base': 1.826233611118671e-05, 'epoch': 0.31}) +[2026-06-29 21:58:31,702][starVLA.training.train_starvla][INFO] - Step 712, Loss: {'train/loss': 0.005826176144182682, 'train/grad_norm_pre_clip': 2.3545303344726562, 'timing/data': 0.0002485790755599737, 'timing/model': 1.231435539899394, 'learning_rate/action_model': 7.298155679924478e-05, 'learning_rate/action_query_proj': 7.298155679924478e-05, 'learning_rate/base': 1.8245389199811196e-05, 'epoch': 0.31}) +[2026-06-29 21:59:04,357][starVLA.training.train_starvla][INFO] - Step 713, Loss: {'train/loss': 0.009994939900934696, 'train/grad_norm_pre_clip': 4.019805431365967, 'timing/data': 0.00032500410452485085, 'timing/model': 1.2346020110417157, 'learning_rate/action_model': 7.291371738383705e-05, 'learning_rate/action_query_proj': 7.291371738383705e-05, 'learning_rate/base': 1.8228429345959263e-05, 'epoch': 0.31}) +[2026-06-29 21:59:37,021][starVLA.training.train_starvla][INFO] - Step 714, Loss: {'train/loss': 0.6515296101570129, 'train/grad_norm_pre_clip': 4.613718509674072, 'timing/data': 0.0002520151901990175, 'timing/model': 1.2287151999771595, 'learning_rate/action_model': 7.284582636591063e-05, 'learning_rate/action_query_proj': 7.284582636591063e-05, 'learning_rate/base': 1.821145659147766e-05, 'epoch': 0.31}) +[2026-06-29 22:00:09,669][starVLA.training.train_starvla][INFO] - Step 715, Loss: {'train/loss': 0.0013120602816343307, 'train/grad_norm_pre_clip': 2.825357437133789, 'timing/data': 0.0002485590521246195, 'timing/model': 1.2297662820201367, 'learning_rate/action_model': 7.277788391297988e-05, 'learning_rate/action_query_proj': 7.277788391297988e-05, 'learning_rate/base': 1.819447097824497e-05, 'epoch': 0.31}) +[2026-06-29 22:00:42,373][starVLA.training.train_starvla][INFO] - Step 716, Loss: {'train/loss': 0.7331854701042175, 'train/grad_norm_pre_clip': 1.8622338771820068, 'timing/data': 0.00026358687318861485, 'timing/model': 1.2373456289060414, 'learning_rate/action_model': 7.270989019268604e-05, 'learning_rate/action_query_proj': 7.270989019268604e-05, 'learning_rate/base': 1.817747254817151e-05, 'epoch': 0.31}) +[2026-06-29 22:01:15,048][starVLA.training.train_starvla][INFO] - Step 717, Loss: {'train/loss': 0.016621463000774384, 'train/grad_norm_pre_clip': 2.002621650695801, 'timing/data': 0.00020676013082265854, 'timing/model': 1.2313456060364842, 'learning_rate/action_model': 7.264184537279687e-05, 'learning_rate/action_query_proj': 7.264184537279687e-05, 'learning_rate/base': 1.8160461343199217e-05, 'epoch': 0.31}) +[2026-06-29 22:01:47,747][starVLA.training.train_starvla][INFO] - Step 718, Loss: {'train/loss': 0.0012018518755212426, 'train/grad_norm_pre_clip': 0.9377778172492981, 'timing/data': 0.0002019810490310192, 'timing/model': 1.231989969033748, 'learning_rate/action_model': 7.257374962120617e-05, 'learning_rate/action_query_proj': 7.257374962120617e-05, 'learning_rate/base': 1.8143437405301543e-05, 'epoch': 0.31}) +[2026-06-29 22:02:20,457][starVLA.training.train_starvla][INFO] - Step 719, Loss: {'train/loss': 0.15083229541778564, 'train/grad_norm_pre_clip': 1.7259738445281982, 'timing/data': 0.0001949579454958439, 'timing/model': 1.2314603270497173, 'learning_rate/action_model': 7.250560310593348e-05, 'learning_rate/action_query_proj': 7.250560310593348e-05, 'learning_rate/base': 1.812640077648337e-05, 'epoch': 0.31}) +[2026-06-29 22:02:53,115][starVLA.training.train_starvla][INFO] - Step 720, Loss: {'train/loss': 0.420790433883667, 'train/grad_norm_pre_clip': 1.826149582862854, 'timing/data': 0.00024602399207651615, 'timing/model': 1.2296958148945123, 'learning_rate/action_model': 7.243740599512349e-05, 'learning_rate/action_query_proj': 7.243740599512349e-05, 'learning_rate/base': 1.8109351498780873e-05, 'epoch': 0.31}) +[2026-06-29 22:03:25,807][starVLA.training.train_starvla][INFO] - Step 721, Loss: {'train/loss': 0.06073610112071037, 'train/grad_norm_pre_clip': 3.3463592529296875, 'timing/data': 0.00024665589444339275, 'timing/model': 1.228669707197696, 'learning_rate/action_model': 7.236915845704585e-05, 'learning_rate/action_query_proj': 7.236915845704585e-05, 'learning_rate/base': 1.8092289614261462e-05, 'epoch': 0.31}) +[2026-06-29 22:03:58,441][starVLA.training.train_starvla][INFO] - Step 722, Loss: {'train/loss': 0.028275195509195328, 'train/grad_norm_pre_clip': 3.46421217918396, 'timing/data': 0.0002367671113461256, 'timing/model': 1.2310632120352238, 'learning_rate/action_model': 7.230086066009455e-05, 'learning_rate/action_query_proj': 7.230086066009455e-05, 'learning_rate/base': 1.8075215165023638e-05, 'epoch': 0.31}) +[2026-06-29 22:04:31,142][starVLA.training.train_starvla][INFO] - Step 723, Loss: {'train/loss': 0.0011613123351708055, 'train/grad_norm_pre_clip': 2.0165622234344482, 'timing/data': 0.0002441809047013521, 'timing/model': 1.2297089849598706, 'learning_rate/action_model': 7.223251277278765e-05, 'learning_rate/action_query_proj': 7.223251277278765e-05, 'learning_rate/base': 1.805812819319691e-05, 'epoch': 0.31}) +[2026-06-29 22:05:03,812][starVLA.training.train_starvla][INFO] - Step 724, Loss: {'train/loss': 0.0002912861236836761, 'train/grad_norm_pre_clip': 1.383749008178711, 'timing/data': 0.00023197801783680916, 'timing/model': 1.2306179581210017, 'learning_rate/action_model': 7.216411496376672e-05, 'learning_rate/action_query_proj': 7.216411496376672e-05, 'learning_rate/base': 1.804102874094168e-05, 'epoch': 0.31}) +[2026-06-29 22:05:36,506][starVLA.training.train_starvla][INFO] - Step 725, Loss: {'train/loss': 0.42946523427963257, 'train/grad_norm_pre_clip': 3.197258949279785, 'timing/data': 0.00024779699742794037, 'timing/model': 1.2275254419073462, 'learning_rate/action_model': 7.209566740179655e-05, 'learning_rate/action_query_proj': 7.209566740179655e-05, 'learning_rate/base': 1.802391685044914e-05, 'epoch': 0.31}) +[2026-06-29 22:06:09,188][starVLA.training.train_starvla][INFO] - Step 726, Loss: {'train/loss': 0.32697635889053345, 'train/grad_norm_pre_clip': 2.0027737617492676, 'timing/data': 0.00025329808704555035, 'timing/model': 1.229561171028763, 'learning_rate/action_model': 7.202717025576475e-05, 'learning_rate/action_query_proj': 7.202717025576475e-05, 'learning_rate/base': 1.8006792563941187e-05, 'epoch': 0.31}) +[2026-06-29 22:06:41,871][starVLA.training.train_starvla][INFO] - Step 727, Loss: {'train/loss': 9.848475019680336e-05, 'train/grad_norm_pre_clip': 3.0074682235717773, 'timing/data': 0.0002581370063126087, 'timing/model': 1.2360124120023102, 'learning_rate/action_model': 7.195862369468117e-05, 'learning_rate/action_query_proj': 7.195862369468117e-05, 'learning_rate/base': 1.798965592367029e-05, 'epoch': 0.31}) +[2026-06-29 22:07:14,557][starVLA.training.train_starvla][INFO] - Step 728, Loss: {'train/loss': 0.04933081194758415, 'train/grad_norm_pre_clip': 1.4940223693847656, 'timing/data': 0.00019874516874551773, 'timing/model': 1.2339574240613729, 'learning_rate/action_model': 7.189002788767765e-05, 'learning_rate/action_query_proj': 7.189002788767765e-05, 'learning_rate/base': 1.797250697191941e-05, 'epoch': 0.31}) +[2026-06-29 22:07:47,278][starVLA.training.train_starvla][INFO] - Step 729, Loss: {'train/loss': 0.0002585427137091756, 'train/grad_norm_pre_clip': 2.0538251399993896, 'timing/data': 0.00019508786499500275, 'timing/model': 1.228697689017281, 'learning_rate/action_model': 7.182138300400751e-05, 'learning_rate/action_query_proj': 7.182138300400751e-05, 'learning_rate/base': 1.7955345751001878e-05, 'epoch': 0.31}) +[2026-06-29 22:08:20,122][starVLA.training.train_starvla][INFO] - Step 730, Loss: {'train/loss': 0.01311985682696104, 'train/grad_norm_pre_clip': 1.2558045387268066, 'timing/data': 0.00029247207567095757, 'timing/model': 1.231962560908869, 'learning_rate/action_model': 7.175268921304524e-05, 'learning_rate/action_query_proj': 7.175268921304524e-05, 'learning_rate/base': 1.793817230326131e-05, 'epoch': 0.31}) +[2026-06-29 22:08:52,756][starVLA.training.train_starvla][INFO] - Step 731, Loss: {'train/loss': 0.0009929921943694353, 'train/grad_norm_pre_clip': 1.9191715717315674, 'timing/data': 0.00030013592913746834, 'timing/model': 1.2347063808701932, 'learning_rate/action_model': 7.168394668428585e-05, 'learning_rate/action_query_proj': 7.168394668428585e-05, 'learning_rate/base': 1.7920986671071463e-05, 'epoch': 0.31}) +[2026-06-29 22:09:25,454][starVLA.training.train_starvla][INFO] - Step 732, Loss: {'train/loss': 0.008988125249743462, 'train/grad_norm_pre_clip': 1.372717022895813, 'timing/data': 0.00019244314171373844, 'timing/model': 1.232698145788163, 'learning_rate/action_model': 7.161515558734476e-05, 'learning_rate/action_query_proj': 7.161515558734476e-05, 'learning_rate/base': 1.790378889683619e-05, 'epoch': 0.31}) +[2026-06-29 22:09:58,106][starVLA.training.train_starvla][INFO] - Step 733, Loss: {'train/loss': 0.28337419033050537, 'train/grad_norm_pre_clip': 2.422246217727661, 'timing/data': 0.00019047013483941555, 'timing/model': 1.2288467830512673, 'learning_rate/action_model': 7.154631609195714e-05, 'learning_rate/action_query_proj': 7.154631609195714e-05, 'learning_rate/base': 1.7886579022989286e-05, 'epoch': 0.32}) +[2026-06-29 22:10:30,849][starVLA.training.train_starvla][INFO] - Step 734, Loss: {'train/loss': 0.0005369113059714437, 'train/grad_norm_pre_clip': 6.03663444519043, 'timing/data': 0.00024966103956103325, 'timing/model': 1.2330574339721352, 'learning_rate/action_model': 7.14774283679776e-05, 'learning_rate/action_query_proj': 7.14774283679776e-05, 'learning_rate/base': 1.78693570919944e-05, 'epoch': 0.32}) +[2026-06-29 22:11:03,587][starVLA.training.train_starvla][INFO] - Step 735, Loss: {'train/loss': 0.006962245795875788, 'train/grad_norm_pre_clip': 1.7363947629928589, 'timing/data': 0.00033256784081459045, 'timing/model': 1.2304383590817451, 'learning_rate/action_model': 7.140849258537977e-05, 'learning_rate/action_query_proj': 7.140849258537977e-05, 'learning_rate/base': 1.7852123146344943e-05, 'epoch': 0.32}) +[2026-06-29 22:11:36,260][starVLA.training.train_starvla][INFO] - Step 736, Loss: {'train/loss': 0.6580316424369812, 'train/grad_norm_pre_clip': 3.211704969406128, 'timing/data': 0.00020718108862638474, 'timing/model': 1.2307388230692595, 'learning_rate/action_model': 7.133950891425582e-05, 'learning_rate/action_query_proj': 7.133950891425582e-05, 'learning_rate/base': 1.7834877228563955e-05, 'epoch': 0.32}) +[2026-06-29 22:12:08,917][starVLA.training.train_starvla][INFO] - Step 737, Loss: {'train/loss': 0.22838681936264038, 'train/grad_norm_pre_clip': 2.994966506958008, 'timing/data': 0.0003932919353246689, 'timing/model': 1.2283472379203886, 'learning_rate/action_model': 7.12704775248161e-05, 'learning_rate/action_query_proj': 7.12704775248161e-05, 'learning_rate/base': 1.7817619381204026e-05, 'epoch': 0.32}) +[2026-06-29 22:12:41,605][starVLA.training.train_starvla][INFO] - Step 738, Loss: {'train/loss': 0.004497454967349768, 'train/grad_norm_pre_clip': 3.2345752716064453, 'timing/data': 0.00020237197168171406, 'timing/model': 1.2318117548711598, 'learning_rate/action_model': 7.12013985873887e-05, 'learning_rate/action_query_proj': 7.12013985873887e-05, 'learning_rate/base': 1.7800349646847176e-05, 'epoch': 0.32}) +[2026-06-29 22:13:14,281][starVLA.training.train_starvla][INFO] - Step 739, Loss: {'train/loss': 0.00514621939510107, 'train/grad_norm_pre_clip': 2.5406839847564697, 'timing/data': 0.0002453029155731201, 'timing/model': 1.2377095939591527, 'learning_rate/action_model': 7.113227227241907e-05, 'learning_rate/action_query_proj': 7.113227227241907e-05, 'learning_rate/base': 1.7783068068104768e-05, 'epoch': 0.32}) +[2026-06-29 22:13:46,958][starVLA.training.train_starvla][INFO] - Step 740, Loss: {'train/loss': 4.738557436212432e-06, 'train/grad_norm_pre_clip': 3.6133646965026855, 'timing/data': 0.00019728182815015316, 'timing/model': 1.2306715629529208, 'learning_rate/action_model': 7.106309875046947e-05, 'learning_rate/action_query_proj': 7.106309875046947e-05, 'learning_rate/base': 1.7765774687617366e-05, 'epoch': 0.32}) +[2026-06-29 22:14:19,687][starVLA.training.train_starvla][INFO] - Step 741, Loss: {'train/loss': 0.06680608540773392, 'train/grad_norm_pre_clip': 1.3253817558288574, 'timing/data': 0.0002816920168697834, 'timing/model': 1.234507581917569, 'learning_rate/action_model': 7.099387819221872e-05, 'learning_rate/action_query_proj': 7.099387819221872e-05, 'learning_rate/base': 1.774846954805468e-05, 'epoch': 0.32}) +[2026-06-29 22:14:52,459][starVLA.training.train_starvla][INFO] - Step 742, Loss: {'train/loss': 7.819453458068892e-05, 'train/grad_norm_pre_clip': 2.784651041030884, 'timing/data': 0.00036202301271259785, 'timing/model': 1.2196549971122295, 'learning_rate/action_model': 7.092461076846166e-05, 'learning_rate/action_query_proj': 7.092461076846166e-05, 'learning_rate/base': 1.7731152692115415e-05, 'epoch': 0.32}) +[2026-06-29 22:15:25,161][starVLA.training.train_starvla][INFO] - Step 743, Loss: {'train/loss': 0.016800515353679657, 'train/grad_norm_pre_clip': 2.6499650478363037, 'timing/data': 0.00025858799926936626, 'timing/model': 1.2309776348993182, 'learning_rate/action_model': 7.085529665010876e-05, 'learning_rate/action_query_proj': 7.085529665010876e-05, 'learning_rate/base': 1.771382416252719e-05, 'epoch': 0.32}) +[2026-06-29 22:15:57,871][starVLA.training.train_starvla][INFO] - Step 744, Loss: {'train/loss': 0.7085421085357666, 'train/grad_norm_pre_clip': 3.6188161373138428, 'timing/data': 0.00019413605332374573, 'timing/model': 1.2268169259186834, 'learning_rate/action_model': 7.078593600818572e-05, 'learning_rate/action_query_proj': 7.078593600818572e-05, 'learning_rate/base': 1.769648400204643e-05, 'epoch': 0.32}) +[2026-06-29 22:16:30,505][starVLA.training.train_starvla][INFO] - Step 745, Loss: {'train/loss': 0.34126368165016174, 'train/grad_norm_pre_clip': 2.344423532485962, 'timing/data': 0.00028017815202474594, 'timing/model': 1.2321649801451713, 'learning_rate/action_model': 7.071652901383304e-05, 'learning_rate/action_query_proj': 7.071652901383304e-05, 'learning_rate/base': 1.767913225345826e-05, 'epoch': 0.32}) +[2026-06-29 22:17:03,156][starVLA.training.train_starvla][INFO] - Step 746, Loss: {'train/loss': 1.8924074538517743e-05, 'train/grad_norm_pre_clip': 2.1942386627197266, 'timing/data': 0.0002481380943208933, 'timing/model': 1.2329064030200243, 'learning_rate/action_model': 7.064707583830558e-05, 'learning_rate/action_query_proj': 7.064707583830558e-05, 'learning_rate/base': 1.7661768959576396e-05, 'epoch': 0.32}) +[2026-06-29 22:17:35,847][starVLA.training.train_starvla][INFO] - Step 747, Loss: {'train/loss': 0.0006226660334505141, 'train/grad_norm_pre_clip': 2.274718761444092, 'timing/data': 0.00025039189495146275, 'timing/model': 1.231747840065509, 'learning_rate/action_model': 7.057757665297213e-05, 'learning_rate/action_query_proj': 7.057757665297213e-05, 'learning_rate/base': 1.7644394163243032e-05, 'epoch': 0.32}) +[2026-06-29 22:18:08,517][starVLA.training.train_starvla][INFO] - Step 748, Loss: {'train/loss': 0.16957876086235046, 'train/grad_norm_pre_clip': 2.227210283279419, 'timing/data': 0.0002857691142708063, 'timing/model': 1.2302464109379798, 'learning_rate/action_model': 7.050803162931504e-05, 'learning_rate/action_query_proj': 7.050803162931504e-05, 'learning_rate/base': 1.762700790732876e-05, 'epoch': 0.32}) +[2026-06-29 22:18:41,129][starVLA.training.train_starvla][INFO] - Step 749, Loss: {'train/loss': 0.003529102075845003, 'train/grad_norm_pre_clip': 1.282920479774475, 'timing/data': 0.0003154550213366747, 'timing/model': 1.2290477650240064, 'learning_rate/action_model': 7.043844093892973e-05, 'learning_rate/action_query_proj': 7.043844093892973e-05, 'learning_rate/base': 1.7609610234732432e-05, 'epoch': 0.32}) +[2026-06-29 22:19:13,824][starVLA.training.train_starvla][INFO] - Step 750, Loss: {'train/loss': 0.0008725962834432721, 'train/grad_norm_pre_clip': 1.77668035030365, 'timing/data': 0.00033921003341674805, 'timing/model': 1.2296517281793058, 'learning_rate/action_model': 7.036880475352432e-05, 'learning_rate/action_query_proj': 7.036880475352432e-05, 'learning_rate/base': 1.759220118838108e-05, 'epoch': 0.32}) +[2026-06-29 22:19:46,516][starVLA.training.train_starvla][INFO] - Step 751, Loss: {'train/loss': 0.2891998887062073, 'train/grad_norm_pre_clip': 3.007850408554077, 'timing/data': 0.0002676749136298895, 'timing/model': 1.2307668279390782, 'learning_rate/action_model': 7.029912324491917e-05, 'learning_rate/action_query_proj': 7.029912324491917e-05, 'learning_rate/base': 1.757478081122979e-05, 'epoch': 0.32}) +[2026-06-29 22:20:19,172][starVLA.training.train_starvla][INFO] - Step 752, Loss: {'train/loss': 0.20361290872097015, 'train/grad_norm_pre_clip': 3.292254686355591, 'timing/data': 0.00019626994617283344, 'timing/model': 1.2349980869330466, 'learning_rate/action_model': 7.022939658504645e-05, 'learning_rate/action_query_proj': 7.022939658504645e-05, 'learning_rate/base': 1.7557349146261614e-05, 'epoch': 0.32}) +[2026-06-29 22:20:51,838][starVLA.training.train_starvla][INFO] - Step 753, Loss: {'train/loss': 0.041560955345630646, 'train/grad_norm_pre_clip': 1.5511714220046997, 'timing/data': 0.00018982798792421818, 'timing/model': 1.2340542620513588, 'learning_rate/action_model': 7.015962494594982e-05, 'learning_rate/action_query_proj': 7.015962494594982e-05, 'learning_rate/base': 1.7539906236487456e-05, 'epoch': 0.32}) +[2026-06-29 22:21:24,525][starVLA.training.train_starvla][INFO] - Step 754, Loss: {'train/loss': 0.27803468704223633, 'train/grad_norm_pre_clip': 2.0763421058654785, 'timing/data': 0.00024089403450489044, 'timing/model': 1.2310541169717908, 'learning_rate/action_model': 7.008980849978382e-05, 'learning_rate/action_query_proj': 7.008980849978382e-05, 'learning_rate/base': 1.7522452124945954e-05, 'epoch': 0.32}) +[2026-06-29 22:21:57,194][starVLA.training.train_starvla][INFO] - Step 755, Loss: {'train/loss': 0.00019523034279700369, 'train/grad_norm_pre_clip': 1.4027512073516846, 'timing/data': 0.0002557528205215931, 'timing/model': 1.2273266601841897, 'learning_rate/action_model': 7.001994741881359e-05, 'learning_rate/action_query_proj': 7.001994741881359e-05, 'learning_rate/base': 1.7504986854703397e-05, 'epoch': 0.32}) +[2026-06-29 22:22:29,885][starVLA.training.train_starvla][INFO] - Step 756, Loss: {'train/loss': 0.15972374379634857, 'train/grad_norm_pre_clip': 4.427134990692139, 'timing/data': 0.0002959480043500662, 'timing/model': 1.229401060147211, 'learning_rate/action_model': 6.995004187541442e-05, 'learning_rate/action_query_proj': 6.995004187541442e-05, 'learning_rate/base': 1.7487510468853606e-05, 'epoch': 0.33}) +[2026-06-29 22:23:02,550][starVLA.training.train_starvla][INFO] - Step 757, Loss: {'train/loss': 0.0030795836355537176, 'train/grad_norm_pre_clip': 3.213700771331787, 'timing/data': 0.0002605919726192951, 'timing/model': 1.231593614909798, 'learning_rate/action_model': 6.98800920420713e-05, 'learning_rate/action_query_proj': 6.98800920420713e-05, 'learning_rate/base': 1.7470023010517825e-05, 'epoch': 0.33}) +[2026-06-29 22:23:35,207][starVLA.training.train_starvla][INFO] - Step 758, Loss: {'train/loss': 0.01445445790886879, 'train/grad_norm_pre_clip': 2.411285400390625, 'timing/data': 0.0003769409377127886, 'timing/model': 1.230531448032707, 'learning_rate/action_model': 6.981009809137846e-05, 'learning_rate/action_query_proj': 6.981009809137846e-05, 'learning_rate/base': 1.7452524522844614e-05, 'epoch': 0.33}) +[2026-06-29 22:24:07,913][starVLA.training.train_starvla][INFO] - Step 759, Loss: {'train/loss': 0.6798177361488342, 'train/grad_norm_pre_clip': 2.347871780395508, 'timing/data': 0.0002769229467958212, 'timing/model': 1.2300477419048548, 'learning_rate/action_model': 6.974006019603902e-05, 'learning_rate/action_query_proj': 6.974006019603902e-05, 'learning_rate/base': 1.7435015049009756e-05, 'epoch': 0.33}) +[2026-06-29 22:24:40,598][starVLA.training.train_starvla][INFO] - Step 760, Loss: {'train/loss': 0.005966988857835531, 'train/grad_norm_pre_clip': 3.2924675941467285, 'timing/data': 0.0002502619754523039, 'timing/model': 1.2286610130686313, 'learning_rate/action_model': 6.966997852886456e-05, 'learning_rate/action_query_proj': 6.966997852886456e-05, 'learning_rate/base': 1.741749463221614e-05, 'epoch': 0.33}) +[2026-06-29 22:25:13,264][starVLA.training.train_starvla][INFO] - Step 761, Loss: {'train/loss': 0.3692893981933594, 'train/grad_norm_pre_clip': 1.9350645542144775, 'timing/data': 0.00023881997913122177, 'timing/model': 1.2272873879410326, 'learning_rate/action_model': 6.959985326277457e-05, 'learning_rate/action_query_proj': 6.959985326277457e-05, 'learning_rate/base': 1.7399963315693643e-05, 'epoch': 0.33}) +[2026-06-29 22:25:45,951][starVLA.training.train_starvla][INFO] - Step 762, Loss: {'train/loss': 0.000470814760774374, 'train/grad_norm_pre_clip': 1.6327296495437622, 'timing/data': 0.00027639209292829037, 'timing/model': 1.2346837408840656, 'learning_rate/action_model': 6.952968457079623e-05, 'learning_rate/action_query_proj': 6.952968457079623e-05, 'learning_rate/base': 1.7382421142699057e-05, 'epoch': 0.33}) +[2026-06-29 22:26:18,602][starVLA.training.train_starvla][INFO] - Step 763, Loss: {'train/loss': 0.00819308590143919, 'train/grad_norm_pre_clip': 3.240814208984375, 'timing/data': 0.0002566440962255001, 'timing/model': 1.2318392121233046, 'learning_rate/action_model': 6.945947262606376e-05, 'learning_rate/action_query_proj': 6.945947262606376e-05, 'learning_rate/base': 1.736486815651594e-05, 'epoch': 0.33}) +[2026-06-29 22:26:51,312][starVLA.training.train_starvla][INFO] - Step 764, Loss: {'train/loss': 0.012463057413697243, 'train/grad_norm_pre_clip': 5.641748905181885, 'timing/data': 0.00020190095528960228, 'timing/model': 1.2334997330326587, 'learning_rate/action_model': 6.93892176018182e-05, 'learning_rate/action_query_proj': 6.93892176018182e-05, 'learning_rate/base': 1.734730440045455e-05, 'epoch': 0.33}) +[2026-06-29 22:27:23,968][starVLA.training.train_starvla][INFO] - Step 765, Loss: {'train/loss': 0.010182990692555904, 'train/grad_norm_pre_clip': 3.462766647338867, 'timing/data': 0.0002505430020391941, 'timing/model': 1.23029455402866, 'learning_rate/action_model': 6.931891967140681e-05, 'learning_rate/action_query_proj': 6.931891967140681e-05, 'learning_rate/base': 1.7329729917851702e-05, 'epoch': 0.33}) +[2026-06-29 22:27:56,629][starVLA.training.train_starvla][INFO] - Step 766, Loss: {'train/loss': 0.018141083419322968, 'train/grad_norm_pre_clip': 1.948052167892456, 'timing/data': 0.00020488700829446316, 'timing/model': 1.2309523930307478, 'learning_rate/action_model': 6.924857900828276e-05, 'learning_rate/action_query_proj': 6.924857900828276e-05, 'learning_rate/base': 1.731214475207069e-05, 'epoch': 0.33}) +[2026-06-29 22:28:29,274][starVLA.training.train_starvla][INFO] - Step 767, Loss: {'train/loss': 0.1161460354924202, 'train/grad_norm_pre_clip': 3.5382080078125, 'timing/data': 0.0002501120325177908, 'timing/model': 1.2331468120682985, 'learning_rate/action_model': 6.917819578600463e-05, 'learning_rate/action_query_proj': 6.917819578600463e-05, 'learning_rate/base': 1.7294548946501158e-05, 'epoch': 0.33}) +[2026-06-29 22:29:01,952][starVLA.training.train_starvla][INFO] - Step 768, Loss: {'train/loss': 0.10222247242927551, 'train/grad_norm_pre_clip': 3.587399959564209, 'timing/data': 0.00020794314332306385, 'timing/model': 1.2282346310094, 'learning_rate/action_model': 6.910777017823603e-05, 'learning_rate/action_query_proj': 6.910777017823603e-05, 'learning_rate/base': 1.727694254455901e-05, 'epoch': 0.33}) +[2026-06-29 22:29:34,670][starVLA.training.train_starvla][INFO] - Step 769, Loss: {'train/loss': 0.7756287455558777, 'train/grad_norm_pre_clip': 4.790502548217773, 'timing/data': 0.00025200587697327137, 'timing/model': 1.2307597040198743, 'learning_rate/action_model': 6.903730235874517e-05, 'learning_rate/action_query_proj': 6.903730235874517e-05, 'learning_rate/base': 1.7259325589686294e-05, 'epoch': 0.33}) +[2026-06-29 22:30:07,347][starVLA.training.train_starvla][INFO] - Step 770, Loss: {'train/loss': 0.0046129385009408, 'train/grad_norm_pre_clip': 3.152782917022705, 'timing/data': 0.000291178934276104, 'timing/model': 1.2341928200330585, 'learning_rate/action_model': 6.896679250140435e-05, 'learning_rate/action_query_proj': 6.896679250140435e-05, 'learning_rate/base': 1.7241698125351087e-05, 'epoch': 0.33}) +[2026-06-29 22:30:40,026][starVLA.training.train_starvla][INFO] - Step 771, Loss: {'train/loss': 0.0068619102239608765, 'train/grad_norm_pre_clip': 1.7073713541030884, 'timing/data': 0.00020359409973025322, 'timing/model': 1.2271660908591002, 'learning_rate/action_model': 6.889624078018966e-05, 'learning_rate/action_query_proj': 6.889624078018966e-05, 'learning_rate/base': 1.7224060195047416e-05, 'epoch': 0.33}) +[2026-06-29 22:31:12,701][starVLA.training.train_starvla][INFO] - Step 772, Loss: {'train/loss': 0.00013228211901150644, 'train/grad_norm_pre_clip': 3.3459830284118652, 'timing/data': 0.00019395607523620129, 'timing/model': 1.2334333050530404, 'learning_rate/action_model': 6.882564736918043e-05, 'learning_rate/action_query_proj': 6.882564736918043e-05, 'learning_rate/base': 1.7206411842295108e-05, 'epoch': 0.33}) +[2026-06-29 22:31:45,345][starVLA.training.train_starvla][INFO] - Step 773, Loss: {'train/loss': 0.0003723702975548804, 'train/grad_norm_pre_clip': 1.2570760250091553, 'timing/data': 0.0001988362055271864, 'timing/model': 1.2304677180945873, 'learning_rate/action_model': 6.875501244255893e-05, 'learning_rate/action_query_proj': 6.875501244255893e-05, 'learning_rate/base': 1.7188753110639734e-05, 'epoch': 0.33}) +[2026-06-29 22:32:18,052][starVLA.training.train_starvla][INFO] - Step 774, Loss: {'train/loss': 0.21191027760505676, 'train/grad_norm_pre_clip': 2.1551828384399414, 'timing/data': 0.00028704083524644375, 'timing/model': 1.2308239999692887, 'learning_rate/action_model': 6.868433617460978e-05, 'learning_rate/action_query_proj': 6.868433617460978e-05, 'learning_rate/base': 1.7171084043652445e-05, 'epoch': 0.33}) +[2026-06-29 22:32:50,685][starVLA.training.train_starvla][INFO] - Step 775, Loss: {'train/loss': 0.6165463328361511, 'train/grad_norm_pre_clip': 3.437295436859131, 'timing/data': 0.00024441094137728214, 'timing/model': 1.219990056939423, 'learning_rate/action_model': 6.861361873971966e-05, 'learning_rate/action_query_proj': 6.861361873971966e-05, 'learning_rate/base': 1.7153404684929915e-05, 'epoch': 0.33}) +[2026-06-29 22:33:23,390][starVLA.training.train_starvla][INFO] - Step 776, Loss: {'train/loss': 0.00490825017914176, 'train/grad_norm_pre_clip': 5.427199363708496, 'timing/data': 0.0002438998781144619, 'timing/model': 1.234596465015784, 'learning_rate/action_model': 6.854286031237682e-05, 'learning_rate/action_query_proj': 6.854286031237682e-05, 'learning_rate/base': 1.7135715078094205e-05, 'epoch': 0.33}) +[2026-06-29 22:33:56,058][starVLA.training.train_starvla][INFO] - Step 777, Loss: {'train/loss': 0.0058822063729166985, 'train/grad_norm_pre_clip': 5.149012565612793, 'timing/data': 0.00019959709607064724, 'timing/model': 1.2272840570658445, 'learning_rate/action_model': 6.847206106717064e-05, 'learning_rate/action_query_proj': 6.847206106717064e-05, 'learning_rate/base': 1.711801526679266e-05, 'epoch': 0.33}) +[2026-06-29 22:34:28,750][starVLA.training.train_starvla][INFO] - Step 778, Loss: {'train/loss': 0.10172971338033676, 'train/grad_norm_pre_clip': 3.4282584190368652, 'timing/data': 0.0002536082174628973, 'timing/model': 1.2332520771306008, 'learning_rate/action_model': 6.840122117879121e-05, 'learning_rate/action_query_proj': 6.840122117879121e-05, 'learning_rate/base': 1.7100305294697802e-05, 'epoch': 0.33}) +[2026-06-29 22:35:01,413][starVLA.training.train_starvla][INFO] - Step 779, Loss: {'train/loss': 0.04098266363143921, 'train/grad_norm_pre_clip': 5.675623893737793, 'timing/data': 0.0001981139648705721, 'timing/model': 1.222669328097254, 'learning_rate/action_model': 6.833034082202892e-05, 'learning_rate/action_query_proj': 6.833034082202892e-05, 'learning_rate/base': 1.708258520550723e-05, 'epoch': 0.34}) +[2026-06-29 22:35:34,104][starVLA.training.train_starvla][INFO] - Step 780, Loss: {'train/loss': 0.02302497997879982, 'train/grad_norm_pre_clip': 2.430727005004883, 'timing/data': 0.00020805303938686848, 'timing/model': 1.230414231074974, 'learning_rate/action_model': 6.8259420171774e-05, 'learning_rate/action_query_proj': 6.8259420171774e-05, 'learning_rate/base': 1.70648550429435e-05, 'epoch': 0.34}) +[2026-06-29 22:36:06,772][starVLA.training.train_starvla][INFO] - Step 781, Loss: {'train/loss': 0.14806385338306427, 'train/grad_norm_pre_clip': 4.223638534545898, 'timing/data': 0.0004366429056972265, 'timing/model': 1.2299016332253814, 'learning_rate/action_model': 6.81884594030161e-05, 'learning_rate/action_query_proj': 6.81884594030161e-05, 'learning_rate/base': 1.7047114850754024e-05, 'epoch': 0.34}) +[2026-06-29 22:36:39,644][starVLA.training.train_starvla][INFO] - Step 782, Loss: {'train/loss': 0.0068941679783165455, 'train/grad_norm_pre_clip': 2.5184781551361084, 'timing/data': 0.00020236102864146233, 'timing/model': 1.2327203401364386, 'learning_rate/action_model': 6.811745869084388e-05, 'learning_rate/action_query_proj': 6.811745869084388e-05, 'learning_rate/base': 1.702936467271097e-05, 'epoch': 0.34}) +[2026-06-29 22:37:12,297][starVLA.training.train_starvla][INFO] - Step 783, Loss: {'train/loss': 0.8345977663993835, 'train/grad_norm_pre_clip': 2.917222261428833, 'timing/data': 0.0001997270155698061, 'timing/model': 1.2287429550196975, 'learning_rate/action_model': 6.804641821044452e-05, 'learning_rate/action_query_proj': 6.804641821044452e-05, 'learning_rate/base': 1.701160455261113e-05, 'epoch': 0.34}) +[2026-06-29 22:37:44,979][starVLA.training.train_starvla][INFO] - Step 784, Loss: {'train/loss': 0.047734204679727554, 'train/grad_norm_pre_clip': 1.4823039770126343, 'timing/data': 0.00026269489899277687, 'timing/model': 1.2296552860643715, 'learning_rate/action_model': 6.797533813710336e-05, 'learning_rate/action_query_proj': 6.797533813710336e-05, 'learning_rate/base': 1.699383453427584e-05, 'epoch': 0.34}) +[2026-06-29 22:38:17,649][starVLA.training.train_starvla][INFO] - Step 785, Loss: {'train/loss': 0.00017187513003591448, 'train/grad_norm_pre_clip': 6.054358005523682, 'timing/data': 0.0003785740118473768, 'timing/model': 1.2294321542140096, 'learning_rate/action_model': 6.790421864620342e-05, 'learning_rate/action_query_proj': 6.790421864620342e-05, 'learning_rate/base': 1.6976054661550854e-05, 'epoch': 0.34}) +[2026-06-29 22:38:50,317][starVLA.training.train_starvla][INFO] - Step 786, Loss: {'train/loss': 0.002726966980844736, 'train/grad_norm_pre_clip': 3.8184213638305664, 'timing/data': 0.000194967957213521, 'timing/model': 1.2299531369935721, 'learning_rate/action_model': 6.783305991322493e-05, 'learning_rate/action_query_proj': 6.783305991322493e-05, 'learning_rate/base': 1.6958264978306233e-05, 'epoch': 0.34}) +[2026-06-29 22:39:22,984][starVLA.training.train_starvla][INFO] - Step 787, Loss: {'train/loss': 0.04098173975944519, 'train/grad_norm_pre_clip': 3.2816038131713867, 'timing/data': 0.0001942068338394165, 'timing/model': 1.2331965470220894, 'learning_rate/action_model': 6.776186211374506e-05, 'learning_rate/action_query_proj': 6.776186211374506e-05, 'learning_rate/base': 1.6940465528436265e-05, 'epoch': 0.34}) +[2026-06-29 22:39:55,638][starVLA.training.train_starvla][INFO] - Step 788, Loss: {'train/loss': 0.0372074730694294, 'train/grad_norm_pre_clip': 3.160120725631714, 'timing/data': 0.00019504781812429428, 'timing/model': 1.231444139033556, 'learning_rate/action_model': 6.769062542343724e-05, 'learning_rate/action_query_proj': 6.769062542343724e-05, 'learning_rate/base': 1.692265635585931e-05, 'epoch': 0.34}) +[2026-06-29 22:40:28,306][starVLA.training.train_starvla][INFO] - Step 789, Loss: {'train/loss': 0.06980227679014206, 'train/grad_norm_pre_clip': 4.125533580780029, 'timing/data': 0.0001991249155253172, 'timing/model': 1.230178751051426, 'learning_rate/action_model': 6.761935001807095e-05, 'learning_rate/action_query_proj': 6.761935001807095e-05, 'learning_rate/base': 1.690483750451774e-05, 'epoch': 0.34}) +[2026-06-29 22:41:00,988][starVLA.training.train_starvla][INFO] - Step 790, Loss: {'train/loss': 0.004530284088104963, 'train/grad_norm_pre_clip': 6.581109046936035, 'timing/data': 0.00020271306857466698, 'timing/model': 1.2308253049850464, 'learning_rate/action_model': 6.754803607351117e-05, 'learning_rate/action_query_proj': 6.754803607351117e-05, 'learning_rate/base': 1.6887009018377793e-05, 'epoch': 0.34}) +[2026-06-29 22:41:33,666][starVLA.training.train_starvla][INFO] - Step 791, Loss: {'train/loss': 0.16344411671161652, 'train/grad_norm_pre_clip': 2.9200868606567383, 'timing/data': 0.00027899607084691525, 'timing/model': 1.2306996290571988, 'learning_rate/action_model': 6.747668376571797e-05, 'learning_rate/action_query_proj': 6.747668376571797e-05, 'learning_rate/base': 1.6869170941429493e-05, 'epoch': 0.34}) +[2026-06-29 22:42:06,349][starVLA.training.train_starvla][INFO] - Step 792, Loss: {'train/loss': 0.23485060036182404, 'train/grad_norm_pre_clip': 2.515901803970337, 'timing/data': 0.00021572713740170002, 'timing/model': 1.2321376509498805, 'learning_rate/action_model': 6.740529327074605e-05, 'learning_rate/action_query_proj': 6.740529327074605e-05, 'learning_rate/base': 1.6851323317686512e-05, 'epoch': 0.34}) +[2026-06-29 22:42:38,982][starVLA.training.train_starvla][INFO] - Step 793, Loss: {'train/loss': 0.149502232670784, 'train/grad_norm_pre_clip': 2.1649181842803955, 'timing/data': 0.0001957588829100132, 'timing/model': 1.231676046969369, 'learning_rate/action_model': 6.73338647647444e-05, 'learning_rate/action_query_proj': 6.73338647647444e-05, 'learning_rate/base': 1.68334661911861e-05, 'epoch': 0.34}) +[2026-06-29 22:43:11,679][starVLA.training.train_starvla][INFO] - Step 794, Loss: {'train/loss': 0.005087490193545818, 'train/grad_norm_pre_clip': 1.4083130359649658, 'timing/data': 0.00019929604604840279, 'timing/model': 1.234480801038444, 'learning_rate/action_model': 6.726239842395571e-05, 'learning_rate/action_query_proj': 6.726239842395571e-05, 'learning_rate/base': 1.681559960598893e-05, 'epoch': 0.34}) +[2026-06-29 22:43:44,356][starVLA.training.train_starvla][INFO] - Step 795, Loss: {'train/loss': 0.00043562351493164897, 'train/grad_norm_pre_clip': 4.915955543518066, 'timing/data': 0.0001959400251507759, 'timing/model': 1.2322714941110462, 'learning_rate/action_model': 6.719089442471615e-05, 'learning_rate/action_query_proj': 6.719089442471615e-05, 'learning_rate/base': 1.6797723606179037e-05, 'epoch': 0.34}) +[2026-06-29 22:44:17,044][starVLA.training.train_starvla][INFO] - Step 796, Loss: {'train/loss': 0.0010863849893212318, 'train/grad_norm_pre_clip': 3.930978536605835, 'timing/data': 0.00036347610875964165, 'timing/model': 1.232634176965803, 'learning_rate/action_model': 6.711935294345463e-05, 'learning_rate/action_query_proj': 6.711935294345463e-05, 'learning_rate/base': 1.6779838235863658e-05, 'epoch': 0.34}) +[2026-06-29 22:44:49,749][starVLA.training.train_starvla][INFO] - Step 797, Loss: {'train/loss': 0.002038255799561739, 'train/grad_norm_pre_clip': 2.567082405090332, 'timing/data': 0.0004715900868177414, 'timing/model': 1.235464271856472, 'learning_rate/action_model': 6.704777415669271e-05, 'learning_rate/action_query_proj': 6.704777415669271e-05, 'learning_rate/base': 1.676194353917318e-05, 'epoch': 0.34}) +[2026-06-29 22:45:22,415][starVLA.training.train_starvla][INFO] - Step 798, Loss: {'train/loss': 0.07228271663188934, 'train/grad_norm_pre_clip': 3.8213953971862793, 'timing/data': 0.00020716991275548935, 'timing/model': 1.2301070201210678, 'learning_rate/action_model': 6.697615824104392e-05, 'learning_rate/action_query_proj': 6.697615824104392e-05, 'learning_rate/base': 1.674403956026098e-05, 'epoch': 0.34}) +[2026-06-29 22:45:55,092][starVLA.training.train_starvla][INFO] - Step 799, Loss: {'train/loss': 0.00033973160316236317, 'train/grad_norm_pre_clip': 3.3728225231170654, 'timing/data': 0.0001913809683173895, 'timing/model': 1.2306414900813252, 'learning_rate/action_model': 6.690450537321339e-05, 'learning_rate/action_query_proj': 6.690450537321339e-05, 'learning_rate/base': 1.6726126343303348e-05, 'epoch': 0.34}) +[2026-06-29 22:46:27,813][starVLA.training.train_starvla][INFO] - Step 800, Loss: {'train/loss': 0.8091009259223938, 'train/grad_norm_pre_clip': 3.670344829559326, 'timing/data': 0.00019979593344032764, 'timing/model': 1.232601447030902, 'learning_rate/action_model': 6.683281572999748e-05, 'learning_rate/action_query_proj': 6.683281572999748e-05, 'learning_rate/base': 1.670820393249937e-05, 'epoch': 0.34}) +[2026-06-29 22:47:00,523][starVLA.training.train_starvla][INFO] - Step 801, Loss: {'train/loss': 0.015064140781760216, 'train/grad_norm_pre_clip': 2.9913501739501953, 'timing/data': 0.00028785294853150845, 'timing/model': 1.2317015880253166, 'learning_rate/action_model': 6.676108948828323e-05, 'learning_rate/action_query_proj': 6.676108948828323e-05, 'learning_rate/base': 1.6690272372070808e-05, 'epoch': 0.34}) +[2026-06-29 22:47:33,226][starVLA.training.train_starvla][INFO] - Step 802, Loss: {'train/loss': 0.001043007243424654, 'train/grad_norm_pre_clip': 1.8173134326934814, 'timing/data': 0.0003005869220942259, 'timing/model': 1.233530004043132, 'learning_rate/action_model': 6.668932682504805e-05, 'learning_rate/action_query_proj': 6.668932682504805e-05, 'learning_rate/base': 1.6672331706262012e-05, 'epoch': 0.35}) +[2026-06-29 22:48:05,939][starVLA.training.train_starvla][INFO] - Step 803, Loss: {'train/loss': 0.0017451138701289892, 'train/grad_norm_pre_clip': 1.6245852708816528, 'timing/data': 0.00019604992121458054, 'timing/model': 1.2333504189737141, 'learning_rate/action_model': 6.661752791735912e-05, 'learning_rate/action_query_proj': 6.661752791735912e-05, 'learning_rate/base': 1.665438197933978e-05, 'epoch': 0.35}) +[2026-06-29 22:48:38,613][starVLA.training.train_starvla][INFO] - Step 804, Loss: {'train/loss': 0.0064512742683291435, 'train/grad_norm_pre_clip': 3.193296194076538, 'timing/data': 0.00029791216365993023, 'timing/model': 1.2337017750833184, 'learning_rate/action_model': 6.654569294237317e-05, 'learning_rate/action_query_proj': 6.654569294237317e-05, 'learning_rate/base': 1.6636423235593292e-05, 'epoch': 0.35}) +[2026-06-29 22:49:11,317][starVLA.training.train_starvla][INFO] - Step 805, Loss: {'train/loss': 0.2509612739086151, 'train/grad_norm_pre_clip': 1.616904377937317, 'timing/data': 0.00032968190498650074, 'timing/model': 1.2323465999215841, 'learning_rate/action_model': 6.64738220773358e-05, 'learning_rate/action_query_proj': 6.64738220773358e-05, 'learning_rate/base': 1.661845551933395e-05, 'epoch': 0.35}) +[2026-06-29 22:49:44,036][starVLA.training.train_starvla][INFO] - Step 806, Loss: {'train/loss': 0.00031045411014929414, 'train/grad_norm_pre_clip': 2.8062241077423096, 'timing/data': 0.0002306150272488594, 'timing/model': 1.2358106249012053, 'learning_rate/action_model': 6.640191549958126e-05, 'learning_rate/action_query_proj': 6.640191549958126e-05, 'learning_rate/base': 1.6600478874895315e-05, 'epoch': 0.35}) +[2026-06-29 22:50:16,717][starVLA.training.train_starvla][INFO] - Step 807, Loss: {'train/loss': 0.09371669590473175, 'train/grad_norm_pre_clip': 1.7611786127090454, 'timing/data': 0.00019245408475399017, 'timing/model': 1.2330513389315456, 'learning_rate/action_model': 6.63299733865319e-05, 'learning_rate/action_query_proj': 6.63299733865319e-05, 'learning_rate/base': 1.6582493346632974e-05, 'epoch': 0.35}) +[2026-06-29 22:50:49,341][starVLA.training.train_starvla][INFO] - Step 808, Loss: {'train/loss': 0.026531772688031197, 'train/grad_norm_pre_clip': 3.569162607192993, 'timing/data': 0.00024030404165387154, 'timing/model': 1.233441422926262, 'learning_rate/action_model': 6.625799591569767e-05, 'learning_rate/action_query_proj': 6.625799591569767e-05, 'learning_rate/base': 1.656449897892442e-05, 'epoch': 0.35}) +[2026-06-29 22:51:22,026][starVLA.training.train_starvla][INFO] - Step 809, Loss: {'train/loss': 0.04170587286353111, 'train/grad_norm_pre_clip': 1.3290942907333374, 'timing/data': 0.0002952469512820244, 'timing/model': 1.231296162120998, 'learning_rate/action_model': 6.61859832646759e-05, 'learning_rate/action_query_proj': 6.61859832646759e-05, 'learning_rate/base': 1.6546495816168975e-05, 'epoch': 0.35}) +[2026-06-29 22:51:54,728][starVLA.training.train_starvla][INFO] - Step 810, Loss: {'train/loss': 0.19160515069961548, 'train/grad_norm_pre_clip': 1.4395650625228882, 'timing/data': 0.00020508700981736183, 'timing/model': 1.2348577871453017, 'learning_rate/action_model': 6.611393561115059e-05, 'learning_rate/action_query_proj': 6.611393561115059e-05, 'learning_rate/base': 1.6528483902787647e-05, 'epoch': 0.35}) +[2026-06-29 22:52:27,436][starVLA.training.train_starvla][INFO] - Step 811, Loss: {'train/loss': 0.0237848162651062, 'train/grad_norm_pre_clip': 2.117100238800049, 'timing/data': 0.00023281900212168694, 'timing/model': 1.2312845310661942, 'learning_rate/action_model': 6.60418531328922e-05, 'learning_rate/action_query_proj': 6.60418531328922e-05, 'learning_rate/base': 1.651046328322305e-05, 'epoch': 0.35}) +[2026-06-29 22:53:00,056][starVLA.training.train_starvla][INFO] - Step 812, Loss: {'train/loss': 7.908870611572638e-05, 'train/grad_norm_pre_clip': 3.2059593200683594, 'timing/data': 0.0003524951171129942, 'timing/model': 1.2301910519599915, 'learning_rate/action_model': 6.596973600775704e-05, 'learning_rate/action_query_proj': 6.596973600775704e-05, 'learning_rate/base': 1.649243400193926e-05, 'epoch': 0.35}) +[2026-06-29 22:53:32,715][starVLA.training.train_starvla][INFO] - Step 813, Loss: {'train/loss': 0.8899930715560913, 'train/grad_norm_pre_clip': 2.1256906986236572, 'timing/data': 0.0001946070697158575, 'timing/model': 1.2339231059886515, 'learning_rate/action_model': 6.589758441368701e-05, 'learning_rate/action_query_proj': 6.589758441368701e-05, 'learning_rate/base': 1.6474396103421752e-05, 'epoch': 0.35}) +[2026-06-29 22:54:05,417][starVLA.training.train_starvla][INFO] - Step 814, Loss: {'train/loss': 0.023716658353805542, 'train/grad_norm_pre_clip': 3.921802282333374, 'timing/data': 0.0002607319038361311, 'timing/model': 1.2344203360844404, 'learning_rate/action_model': 6.582539852870894e-05, 'learning_rate/action_query_proj': 6.582539852870894e-05, 'learning_rate/base': 1.6456349632177234e-05, 'epoch': 0.35}) +[2026-06-29 22:54:38,112][starVLA.training.train_starvla][INFO] - Step 815, Loss: {'train/loss': 0.00013189337914809585, 'train/grad_norm_pre_clip': 2.4628329277038574, 'timing/data': 0.0002277800813317299, 'timing/model': 1.2300964798778296, 'learning_rate/action_model': 6.575317853093433e-05, 'learning_rate/action_query_proj': 6.575317853093433e-05, 'learning_rate/base': 1.6438294632733583e-05, 'epoch': 0.35}) +[2026-06-29 22:55:10,803][starVLA.training.train_starvla][INFO] - Step 816, Loss: {'train/loss': 0.003472347976639867, 'train/grad_norm_pre_clip': 2.7194831371307373, 'timing/data': 0.0001906000543385744, 'timing/model': 1.2340621831826866, 'learning_rate/action_model': 6.568092459855886e-05, 'learning_rate/action_query_proj': 6.568092459855886e-05, 'learning_rate/base': 1.6420231149639714e-05, 'epoch': 0.35}) +[2026-06-29 22:55:43,461][starVLA.training.train_starvla][INFO] - Step 817, Loss: {'train/loss': 0.0066040148958563805, 'train/grad_norm_pre_clip': 2.3905041217803955, 'timing/data': 0.0001932040322571993, 'timing/model': 1.230675534112379, 'learning_rate/action_model': 6.560863690986193e-05, 'learning_rate/action_query_proj': 6.560863690986193e-05, 'learning_rate/base': 1.6402159227465482e-05, 'epoch': 0.35}) +[2026-06-29 22:56:16,088][starVLA.training.train_starvla][INFO] - Step 818, Loss: {'train/loss': 0.030690571293234825, 'train/grad_norm_pre_clip': 1.3487582206726074, 'timing/data': 0.00023253797553479671, 'timing/model': 1.2305838190950453, 'learning_rate/action_model': 6.553631564320621e-05, 'learning_rate/action_query_proj': 6.553631564320621e-05, 'learning_rate/base': 1.6384078910801553e-05, 'epoch': 0.35}) +[2026-06-29 22:56:48,769][starVLA.training.train_starvla][INFO] - Step 819, Loss: {'train/loss': 0.00020586908794939518, 'train/grad_norm_pre_clip': 4.345595836639404, 'timing/data': 0.0004626929294317961, 'timing/model': 1.233445568010211, 'learning_rate/action_model': 6.546396097703726e-05, 'learning_rate/action_query_proj': 6.546396097703726e-05, 'learning_rate/base': 1.6365990244259316e-05, 'epoch': 0.35}) +[2026-06-29 22:57:21,490][starVLA.training.train_starvla][INFO] - Step 820, Loss: {'train/loss': 0.1869632452726364, 'train/grad_norm_pre_clip': 3.3882386684417725, 'timing/data': 0.00019719195552170277, 'timing/model': 1.233741895062849, 'learning_rate/action_model': 6.539157308988302e-05, 'learning_rate/action_query_proj': 6.539157308988302e-05, 'learning_rate/base': 1.6347893272470754e-05, 'epoch': 0.35}) +[2026-06-29 22:57:54,180][starVLA.training.train_starvla][INFO] - Step 821, Loss: {'train/loss': 0.012430922128260136, 'train/grad_norm_pre_clip': 2.0955734252929688, 'timing/data': 0.00029328279197216034, 'timing/model': 1.2296098368242383, 'learning_rate/action_model': 6.531915216035338e-05, 'learning_rate/action_query_proj': 6.531915216035338e-05, 'learning_rate/base': 1.6329788040088346e-05, 'epoch': 0.35}) +[2026-06-29 22:58:26,853][starVLA.training.train_starvla][INFO] - Step 822, Loss: {'train/loss': 0.005369663704186678, 'train/grad_norm_pre_clip': 1.758310317993164, 'timing/data': 0.0002661121543496847, 'timing/model': 1.2336941401008517, 'learning_rate/action_model': 6.524669836713981e-05, 'learning_rate/action_query_proj': 6.524669836713981e-05, 'learning_rate/base': 1.6311674591784953e-05, 'epoch': 0.35}) +[2026-06-29 22:58:59,536][starVLA.training.train_starvla][INFO] - Step 823, Loss: {'train/loss': 0.3830805718898773, 'train/grad_norm_pre_clip': 3.5928919315338135, 'timing/data': 0.000193354906514287, 'timing/model': 1.225660591153428, 'learning_rate/action_model': 6.517421188901482e-05, 'learning_rate/action_query_proj': 6.517421188901482e-05, 'learning_rate/base': 1.6293552972253705e-05, 'epoch': 0.35}) +[2026-06-29 22:59:32,200][starVLA.training.train_starvla][INFO] - Step 824, Loss: {'train/loss': 0.0032024350948631763, 'train/grad_norm_pre_clip': 2.3788695335388184, 'timing/data': 0.000254268990829587, 'timing/model': 1.2338144700042903, 'learning_rate/action_model': 6.510169290483163e-05, 'learning_rate/action_query_proj': 6.510169290483163e-05, 'learning_rate/base': 1.6275423226207907e-05, 'epoch': 0.35}) +[2026-06-29 23:00:04,899][starVLA.training.train_starvla][INFO] - Step 825, Loss: {'train/loss': 0.04644766077399254, 'train/grad_norm_pre_clip': 2.1114420890808105, 'timing/data': 0.0001984338741749525, 'timing/model': 1.2274914069566876, 'learning_rate/action_model': 6.502914159352357e-05, 'learning_rate/action_query_proj': 6.502914159352357e-05, 'learning_rate/base': 1.6257285398380893e-05, 'epoch': 0.35}) +[2026-06-29 23:00:37,574][starVLA.training.train_starvla][INFO] - Step 826, Loss: {'train/loss': 0.004618800710886717, 'train/grad_norm_pre_clip': 2.338721990585327, 'timing/data': 0.0002525260206311941, 'timing/model': 1.2308001250494272, 'learning_rate/action_model': 6.495655813410384e-05, 'learning_rate/action_query_proj': 6.495655813410384e-05, 'learning_rate/base': 1.623913953352596e-05, 'epoch': 0.36}) +[2026-06-29 23:01:10,277][starVLA.training.train_starvla][INFO] - Step 827, Loss: {'train/loss': 0.007760991808027029, 'train/grad_norm_pre_clip': 1.9729706048965454, 'timing/data': 0.0002730349078774452, 'timing/model': 1.2321638830471784, 'learning_rate/action_model': 6.488394270566487e-05, 'learning_rate/action_query_proj': 6.488394270566487e-05, 'learning_rate/base': 1.6220985676416217e-05, 'epoch': 0.36}) +[2026-06-29 23:01:42,947][starVLA.training.train_starvla][INFO] - Step 828, Loss: {'train/loss': 0.005650234408676624, 'train/grad_norm_pre_clip': 1.616519808769226, 'timing/data': 0.00026049092411994934, 'timing/model': 1.2299886350519955, 'learning_rate/action_model': 6.481129548737803e-05, 'learning_rate/action_query_proj': 6.481129548737803e-05, 'learning_rate/base': 1.6202823871844507e-05, 'epoch': 0.36}) +[2026-06-29 23:02:15,607][starVLA.training.train_starvla][INFO] - Step 829, Loss: {'train/loss': 0.029890548437833786, 'train/grad_norm_pre_clip': 1.5214651823043823, 'timing/data': 0.00019772304221987724, 'timing/model': 1.2316032350063324, 'learning_rate/action_model': 6.473861665849312e-05, 'learning_rate/action_query_proj': 6.473861665849312e-05, 'learning_rate/base': 1.618465416462328e-05, 'epoch': 0.36}) +[2026-06-29 23:02:48,309][starVLA.training.train_starvla][INFO] - Step 830, Loss: {'train/loss': 0.2688521146774292, 'train/grad_norm_pre_clip': 1.6392155885696411, 'timing/data': 0.00026329606771469116, 'timing/model': 1.2261871108785272, 'learning_rate/action_model': 6.46659063983379e-05, 'learning_rate/action_query_proj': 6.46659063983379e-05, 'learning_rate/base': 1.6166476599584475e-05, 'epoch': 0.36}) +[2026-06-29 23:03:21,012][starVLA.training.train_starvla][INFO] - Step 831, Loss: {'train/loss': 0.0038805068470537663, 'train/grad_norm_pre_clip': 2.344675302505493, 'timing/data': 0.00025382894091308117, 'timing/model': 1.233165921876207, 'learning_rate/action_model': 6.459316488631774e-05, 'learning_rate/action_query_proj': 6.459316488631774e-05, 'learning_rate/base': 1.6148291221579436e-05, 'epoch': 0.36}) +[2026-06-29 23:03:53,679][starVLA.training.train_starvla][INFO] - Step 832, Loss: {'train/loss': 0.003876582020893693, 'train/grad_norm_pre_clip': 2.348342180252075, 'timing/data': 0.0002616338897496462, 'timing/model': 1.2307275659404695, 'learning_rate/action_model': 6.452039230191506e-05, 'learning_rate/action_query_proj': 6.452039230191506e-05, 'learning_rate/base': 1.6130098075478765e-05, 'epoch': 0.36}) +[2026-06-29 23:04:26,373][starVLA.training.train_starvla][INFO] - Step 833, Loss: {'train/loss': 0.44107961654663086, 'train/grad_norm_pre_clip': 1.3898905515670776, 'timing/data': 0.00021989503875374794, 'timing/model': 1.2314785858616233, 'learning_rate/action_model': 6.444758882468901e-05, 'learning_rate/action_query_proj': 6.444758882468901e-05, 'learning_rate/base': 1.6111897206172254e-05, 'epoch': 0.36}) +[2026-06-29 23:04:59,265][starVLA.training.train_starvla][INFO] - Step 834, Loss: {'train/loss': 0.010227187536656857, 'train/grad_norm_pre_clip': 1.422174334526062, 'timing/data': 0.00019483687356114388, 'timing/model': 1.2366361198946834, 'learning_rate/action_model': 6.437475463427489e-05, 'learning_rate/action_query_proj': 6.437475463427489e-05, 'learning_rate/base': 1.609368865856872e-05, 'epoch': 0.36}) +[2026-06-29 23:05:31,965][starVLA.training.train_starvla][INFO] - Step 835, Loss: {'train/loss': 0.004941241815686226, 'train/grad_norm_pre_clip': 1.913732886314392, 'timing/data': 0.00030276086181402206, 'timing/model': 1.222179318079725, 'learning_rate/action_model': 6.430188991038385e-05, 'learning_rate/action_query_proj': 6.430188991038385e-05, 'learning_rate/base': 1.6075472477595962e-05, 'epoch': 0.36}) +[2026-06-29 23:06:04,645][starVLA.training.train_starvla][INFO] - Step 836, Loss: {'train/loss': 0.29028335213661194, 'train/grad_norm_pre_clip': 1.8683078289031982, 'timing/data': 0.00020044785924255848, 'timing/model': 1.231906812870875, 'learning_rate/action_model': 6.422899483280234e-05, 'learning_rate/action_query_proj': 6.422899483280234e-05, 'learning_rate/base': 1.6057248708200585e-05, 'epoch': 0.36}) +[2026-06-29 23:06:37,317][starVLA.training.train_starvla][INFO] - Step 837, Loss: {'train/loss': 0.08934404700994492, 'train/grad_norm_pre_clip': 1.6289716958999634, 'timing/data': 0.00025291694328188896, 'timing/model': 1.228648599004373, 'learning_rate/action_model': 6.415606958139176e-05, 'learning_rate/action_query_proj': 6.415606958139176e-05, 'learning_rate/base': 1.603901739534794e-05, 'epoch': 0.36}) +[2026-06-29 23:07:09,996][starVLA.training.train_starvla][INFO] - Step 838, Loss: {'train/loss': 0.000570447591599077, 'train/grad_norm_pre_clip': 1.9434125423431396, 'timing/data': 0.0002494710497558117, 'timing/model': 1.2300945702008903, 'learning_rate/action_model': 6.408311433608787e-05, 'learning_rate/action_query_proj': 6.408311433608787e-05, 'learning_rate/base': 1.6020778584021967e-05, 'epoch': 0.36}) +[2026-06-29 23:07:42,700][starVLA.training.train_starvla][INFO] - Step 839, Loss: {'train/loss': 0.1732388436794281, 'train/grad_norm_pre_clip': 1.8132096529006958, 'timing/data': 0.000382482074201107, 'timing/model': 1.2335280999541283, 'learning_rate/action_model': 6.40101292769005e-05, 'learning_rate/action_query_proj': 6.40101292769005e-05, 'learning_rate/base': 1.6002532319225124e-05, 'epoch': 0.36}) +[2026-06-29 23:08:15,367][starVLA.training.train_starvla][INFO] - Step 840, Loss: {'train/loss': 0.04455266892910004, 'train/grad_norm_pre_clip': 1.8750454187393188, 'timing/data': 0.0002027719747275114, 'timing/model': 1.2308475759346038, 'learning_rate/action_model': 6.393711458391304e-05, 'learning_rate/action_query_proj': 6.393711458391304e-05, 'learning_rate/base': 1.598427864597826e-05, 'epoch': 0.36}) +[2026-06-29 23:08:48,056][starVLA.training.train_starvla][INFO] - Step 841, Loss: {'train/loss': 0.06282015889883041, 'train/grad_norm_pre_clip': 1.9957448244094849, 'timing/data': 0.0003963378258049488, 'timing/model': 1.229897283948958, 'learning_rate/action_model': 6.386407043728197e-05, 'learning_rate/action_query_proj': 6.386407043728197e-05, 'learning_rate/base': 1.5966017609320492e-05, 'epoch': 0.36}) +[2026-06-29 23:09:20,734][starVLA.training.train_starvla][INFO] - Step 842, Loss: {'train/loss': 0.0011009280569851398, 'train/grad_norm_pre_clip': 2.604245901107788, 'timing/data': 0.00020501692779362202, 'timing/model': 1.2310946360230446, 'learning_rate/action_model': 6.379099701723648e-05, 'learning_rate/action_query_proj': 6.379099701723648e-05, 'learning_rate/base': 1.594774925430912e-05, 'epoch': 0.36}) +[2026-06-29 23:09:53,431][starVLA.training.train_starvla][INFO] - Step 843, Loss: {'train/loss': 0.040304653346538544, 'train/grad_norm_pre_clip': 4.759143829345703, 'timing/data': 0.0003144829533994198, 'timing/model': 1.2308218420948833, 'learning_rate/action_model': 6.371789450407795e-05, 'learning_rate/action_query_proj': 6.371789450407795e-05, 'learning_rate/base': 1.5929473626019486e-05, 'epoch': 0.36}) +[2026-06-29 23:10:26,105][starVLA.training.train_starvla][INFO] - Step 844, Loss: {'train/loss': 0.0001991825265577063, 'train/grad_norm_pre_clip': 2.3009190559387207, 'timing/data': 0.00025301799178123474, 'timing/model': 1.229416151996702, 'learning_rate/action_model': 6.364476307817956e-05, 'learning_rate/action_query_proj': 6.364476307817956e-05, 'learning_rate/base': 1.591119076954489e-05, 'epoch': 0.36}) +[2026-06-29 23:10:58,751][starVLA.training.train_starvla][INFO] - Step 845, Loss: {'train/loss': 0.01855754666030407, 'train/grad_norm_pre_clip': 2.2987210750579834, 'timing/data': 0.0002482288982719183, 'timing/model': 1.2256154930219054, 'learning_rate/action_model': 6.357160291998586e-05, 'learning_rate/action_query_proj': 6.357160291998586e-05, 'learning_rate/base': 1.5892900729996465e-05, 'epoch': 0.36}) +[2026-06-29 23:11:31,448][starVLA.training.train_starvla][INFO] - Step 846, Loss: {'train/loss': 0.000655073206871748, 'train/grad_norm_pre_clip': 0.9874114990234375, 'timing/data': 0.0001929339487105608, 'timing/model': 1.2409396281000227, 'learning_rate/action_model': 6.349841421001226e-05, 'learning_rate/action_query_proj': 6.349841421001226e-05, 'learning_rate/base': 1.5874603552503065e-05, 'epoch': 0.36}) +[2026-06-29 23:12:04,165][starVLA.training.train_starvla][INFO] - Step 847, Loss: {'train/loss': 0.3008926212787628, 'train/grad_norm_pre_clip': 3.4783051013946533, 'timing/data': 0.0002544401213526726, 'timing/model': 1.2358443960547447, 'learning_rate/action_model': 6.34251971288446e-05, 'learning_rate/action_query_proj': 6.34251971288446e-05, 'learning_rate/base': 1.585629928221115e-05, 'epoch': 0.36}) +[2026-06-29 23:12:36,886][starVLA.training.train_starvla][INFO] - Step 848, Loss: {'train/loss': 0.07554052770137787, 'train/grad_norm_pre_clip': 1.2809579372406006, 'timing/data': 0.0003670628648251295, 'timing/model': 1.2336582629941404, 'learning_rate/action_model': 6.335195185713879e-05, 'learning_rate/action_query_proj': 6.335195185713879e-05, 'learning_rate/base': 1.5837987964284697e-05, 'epoch': 0.36}) +[2026-06-29 23:13:09,569][starVLA.training.train_starvla][INFO] - Step 849, Loss: {'train/loss': 0.042323242872953415, 'train/grad_norm_pre_clip': 2.271580696105957, 'timing/data': 0.00024046399630606174, 'timing/model': 1.2353739398531616, 'learning_rate/action_model': 6.327867857562023e-05, 'learning_rate/action_query_proj': 6.327867857562023e-05, 'learning_rate/base': 1.5819669643905057e-05, 'epoch': 0.37}) +[2026-06-29 23:13:42,292][starVLA.training.train_starvla][INFO] - Step 850, Loss: {'train/loss': 0.029066698625683784, 'train/grad_norm_pre_clip': 2.566802740097046, 'timing/data': 0.00033807801082730293, 'timing/model': 1.2288218589965254, 'learning_rate/action_model': 6.320537746508347e-05, 'learning_rate/action_query_proj': 6.320537746508347e-05, 'learning_rate/base': 1.5801344366270868e-05, 'epoch': 0.37}) +[2026-06-29 23:14:14,994][starVLA.training.train_starvla][INFO] - Step 851, Loss: {'train/loss': 0.05266805738210678, 'train/grad_norm_pre_clip': 1.4021011590957642, 'timing/data': 0.0002651198301464319, 'timing/model': 1.2327012699097395, 'learning_rate/action_model': 6.313204870639171e-05, 'learning_rate/action_query_proj': 6.313204870639171e-05, 'learning_rate/base': 1.5783012176597927e-05, 'epoch': 0.37}) +[2026-06-29 23:14:47,703][starVLA.training.train_starvla][INFO] - Step 852, Loss: {'train/loss': 0.33856719732284546, 'train/grad_norm_pre_clip': 2.9301700592041016, 'timing/data': 0.0002694090362638235, 'timing/model': 1.2314173891209066, 'learning_rate/action_model': 6.305869248047635e-05, 'learning_rate/action_query_proj': 6.305869248047635e-05, 'learning_rate/base': 1.576467312011909e-05, 'epoch': 0.37}) +[2026-06-29 23:15:20,376][starVLA.training.train_starvla][INFO] - Step 853, Loss: {'train/loss': 0.03998535871505737, 'train/grad_norm_pre_clip': 3.979137659072876, 'timing/data': 0.0002528571058064699, 'timing/model': 1.2344331131316721, 'learning_rate/action_model': 6.298530896833663e-05, 'learning_rate/action_query_proj': 6.298530896833663e-05, 'learning_rate/base': 1.5746327242084157e-05, 'epoch': 0.37}) +[2026-06-29 23:15:53,072][starVLA.training.train_starvla][INFO] - Step 854, Loss: {'train/loss': 0.07958105206489563, 'train/grad_norm_pre_clip': 3.1233572959899902, 'timing/data': 0.00023450306616723537, 'timing/model': 1.2340775590855628, 'learning_rate/action_model': 6.291189835103903e-05, 'learning_rate/action_query_proj': 6.291189835103903e-05, 'learning_rate/base': 1.572797458775976e-05, 'epoch': 0.37}) +[2026-06-29 23:16:25,775][starVLA.training.train_starvla][INFO] - Step 855, Loss: {'train/loss': 3.65658852388151e-05, 'train/grad_norm_pre_clip': 0.6098530888557434, 'timing/data': 0.00022001517936587334, 'timing/model': 1.2338241250254214, 'learning_rate/action_model': 6.283846080971699e-05, 'learning_rate/action_query_proj': 6.283846080971699e-05, 'learning_rate/base': 1.5709615202429247e-05, 'epoch': 0.37}) +[2026-06-29 23:16:58,490][starVLA.training.train_starvla][INFO] - Step 856, Loss: {'train/loss': 0.00039753603050485253, 'train/grad_norm_pre_clip': 2.627903938293457, 'timing/data': 0.00019293418154120445, 'timing/model': 1.2364854600746185, 'learning_rate/action_model': 6.276499652557031e-05, 'learning_rate/action_query_proj': 6.276499652557031e-05, 'learning_rate/base': 1.5691249131392578e-05, 'epoch': 0.37}) +[2026-06-29 23:17:31,227][starVLA.training.train_starvla][INFO] - Step 857, Loss: {'train/loss': 0.005411230958998203, 'train/grad_norm_pre_clip': 1.5704596042633057, 'timing/data': 0.00026707397773861885, 'timing/model': 1.2421004949137568, 'learning_rate/action_model': 6.269150567986482e-05, 'learning_rate/action_query_proj': 6.269150567986482e-05, 'learning_rate/base': 1.5672876419966204e-05, 'epoch': 0.37}) +[2026-06-29 23:18:03,931][starVLA.training.train_starvla][INFO] - Step 858, Loss: {'train/loss': 0.0003934933920390904, 'train/grad_norm_pre_clip': 3.4325203895568848, 'timing/data': 0.000291039003059268, 'timing/model': 1.231581622036174, 'learning_rate/action_model': 6.261798845393188e-05, 'learning_rate/action_query_proj': 6.261798845393188e-05, 'learning_rate/base': 1.565449711348297e-05, 'epoch': 0.37}) +[2026-06-29 23:18:36,605][starVLA.training.train_starvla][INFO] - Step 859, Loss: {'train/loss': 0.021902069449424744, 'train/grad_norm_pre_clip': 2.2497217655181885, 'timing/data': 0.00019402592442929745, 'timing/model': 1.2280845360364765, 'learning_rate/action_model': 6.254444502916793e-05, 'learning_rate/action_query_proj': 6.254444502916793e-05, 'learning_rate/base': 1.563611125729198e-05, 'epoch': 0.37}) +[2026-06-29 23:19:09,318][starVLA.training.train_starvla][INFO] - Step 860, Loss: {'train/loss': 0.0006687985733151436, 'train/grad_norm_pre_clip': 2.3826982975006104, 'timing/data': 0.0001969621516764164, 'timing/model': 1.231622370891273, 'learning_rate/action_model': 6.247087558703405e-05, 'learning_rate/action_query_proj': 6.247087558703405e-05, 'learning_rate/base': 1.561771889675851e-05, 'epoch': 0.37}) +[2026-06-29 23:19:41,996][starVLA.training.train_starvla][INFO] - Step 861, Loss: {'train/loss': 0.09754203259944916, 'train/grad_norm_pre_clip': 2.158996820449829, 'timing/data': 0.00025003193877637386, 'timing/model': 1.2322304979898036, 'learning_rate/action_model': 6.239728030905555e-05, 'learning_rate/action_query_proj': 6.239728030905555e-05, 'learning_rate/base': 1.559932007726389e-05, 'epoch': 0.37}) +[2026-06-29 23:20:14,723][starVLA.training.train_starvla][INFO] - Step 862, Loss: {'train/loss': 0.09543181210756302, 'train/grad_norm_pre_clip': 3.460421562194824, 'timing/data': 0.00033661513589322567, 'timing/model': 1.226679515093565, 'learning_rate/action_model': 6.232365937682145e-05, 'learning_rate/action_query_proj': 6.232365937682145e-05, 'learning_rate/base': 1.558091484420536e-05, 'epoch': 0.37}) +[2026-06-29 23:20:47,398][starVLA.training.train_starvla][INFO] - Step 863, Loss: {'train/loss': 0.00542336655780673, 'train/grad_norm_pre_clip': 1.7713472843170166, 'timing/data': 0.0002881130203604698, 'timing/model': 1.2307328910101205, 'learning_rate/action_model': 6.225001297198408e-05, 'learning_rate/action_query_proj': 6.225001297198408e-05, 'learning_rate/base': 1.556250324299602e-05, 'epoch': 0.37}) +[2026-06-29 23:21:20,075][starVLA.training.train_starvla][INFO] - Step 864, Loss: {'train/loss': 0.30168843269348145, 'train/grad_norm_pre_clip': 2.253748893737793, 'timing/data': 0.0002619049046188593, 'timing/model': 1.233697511954233, 'learning_rate/action_model': 6.217634127625863e-05, 'learning_rate/action_query_proj': 6.217634127625863e-05, 'learning_rate/base': 1.5544085319064658e-05, 'epoch': 0.37}) +[2026-06-29 23:21:52,771][starVLA.training.train_starvla][INFO] - Step 865, Loss: {'train/loss': 0.01165467593818903, 'train/grad_norm_pre_clip': 3.651055335998535, 'timing/data': 0.0002571151126176119, 'timing/model': 1.2299571728799492, 'learning_rate/action_model': 6.210264447142266e-05, 'learning_rate/action_query_proj': 6.210264447142266e-05, 'learning_rate/base': 1.5525661117855666e-05, 'epoch': 0.37}) +[2026-06-29 23:22:25,414][starVLA.training.train_starvla][INFO] - Step 866, Loss: {'train/loss': 0.06569620221853256, 'train/grad_norm_pre_clip': 2.25551176071167, 'timing/data': 0.0002852978650480509, 'timing/model': 1.231292018899694, 'learning_rate/action_model': 6.202892273931577e-05, 'learning_rate/action_query_proj': 6.202892273931577e-05, 'learning_rate/base': 1.550723068482894e-05, 'epoch': 0.37}) +[2026-06-29 23:22:58,088][starVLA.training.train_starvla][INFO] - Step 867, Loss: {'train/loss': 8.007413998711854e-05, 'train/grad_norm_pre_clip': 2.372535467147827, 'timing/data': 0.00025344802998006344, 'timing/model': 1.2276864831801504, 'learning_rate/action_model': 6.195517626183894e-05, 'learning_rate/action_query_proj': 6.195517626183894e-05, 'learning_rate/base': 1.5488794065459736e-05, 'epoch': 0.37}) +[2026-06-29 23:23:30,759][starVLA.training.train_starvla][INFO] - Step 868, Loss: {'train/loss': 0.00956137664616108, 'train/grad_norm_pre_clip': 2.7387304306030273, 'timing/data': 0.0001949670258909464, 'timing/model': 1.231092233909294, 'learning_rate/action_model': 6.188140522095431e-05, 'learning_rate/action_query_proj': 6.188140522095431e-05, 'learning_rate/base': 1.5470351305238578e-05, 'epoch': 0.37}) +[2026-06-29 23:24:03,452][starVLA.training.train_starvla][INFO] - Step 869, Loss: {'train/loss': 0.0006836071843281388, 'train/grad_norm_pre_clip': 1.7955678701400757, 'timing/data': 0.00028567900881171227, 'timing/model': 1.2353986860252917, 'learning_rate/action_model': 6.180760979868458e-05, 'learning_rate/action_query_proj': 6.180760979868458e-05, 'learning_rate/base': 1.5451902449671146e-05, 'epoch': 0.37}) +[2026-06-29 23:24:36,048][starVLA.training.train_starvla][INFO] - Step 870, Loss: {'train/loss': 0.00014802867372054607, 'train/grad_norm_pre_clip': 1.6483420133590698, 'timing/data': 0.00025361799634993076, 'timing/model': 1.2322986421640962, 'learning_rate/action_model': 6.17337901771126e-05, 'learning_rate/action_query_proj': 6.17337901771126e-05, 'learning_rate/base': 1.543344754427815e-05, 'epoch': 0.37}) +[2026-06-29 23:25:08,768][starVLA.training.train_starvla][INFO] - Step 871, Loss: {'train/loss': 0.0326797291636467, 'train/grad_norm_pre_clip': 1.8609659671783447, 'timing/data': 0.0001905001699924469, 'timing/model': 1.238323133904487, 'learning_rate/action_model': 6.165994653838099e-05, 'learning_rate/action_query_proj': 6.165994653838099e-05, 'learning_rate/base': 1.5414986634595247e-05, 'epoch': 0.37}) +[2026-06-29 23:25:41,433][starVLA.training.train_starvla][INFO] - Step 872, Loss: {'train/loss': 0.004212557803839445, 'train/grad_norm_pre_clip': 3.272388458251953, 'timing/data': 0.0002052269410341978, 'timing/model': 1.2324418420903385, 'learning_rate/action_model': 6.158607906469154e-05, 'learning_rate/action_query_proj': 6.158607906469154e-05, 'learning_rate/base': 1.5396519766172884e-05, 'epoch': 0.38}) +[2026-06-29 23:26:14,136][starVLA.training.train_starvla][INFO] - Step 873, Loss: {'train/loss': 0.00036552222445607185, 'train/grad_norm_pre_clip': 1.8095977306365967, 'timing/data': 0.00035231513902544975, 'timing/model': 1.2352012139745057, 'learning_rate/action_model': 6.151218793830495e-05, 'learning_rate/action_query_proj': 6.151218793830495e-05, 'learning_rate/base': 1.5378046984576238e-05, 'epoch': 0.38}) +[2026-06-29 23:26:46,737][starVLA.training.train_starvla][INFO] - Step 874, Loss: {'train/loss': 0.052611611783504486, 'train/grad_norm_pre_clip': 2.297853946685791, 'timing/data': 0.00025282707065343857, 'timing/model': 1.2246786651667207, 'learning_rate/action_model': 6.143827334154016e-05, 'learning_rate/action_query_proj': 6.143827334154016e-05, 'learning_rate/base': 1.535956833538504e-05, 'epoch': 0.38}) +[2026-06-29 23:27:19,377][starVLA.training.train_starvla][INFO] - Step 875, Loss: {'train/loss': 0.000332154450006783, 'train/grad_norm_pre_clip': 1.7512571811676025, 'timing/data': 0.00028824410401284695, 'timing/model': 1.2294202549383044, 'learning_rate/action_model': 6.136433545677417e-05, 'learning_rate/action_query_proj': 6.136433545677417e-05, 'learning_rate/base': 1.534108386419354e-05, 'epoch': 0.38}) +[2026-06-29 23:27:52,064][starVLA.training.train_starvla][INFO] - Step 876, Loss: {'train/loss': 8.058003732003272e-05, 'train/grad_norm_pre_clip': 4.289503574371338, 'timing/data': 0.0002544000744819641, 'timing/model': 1.2322211530990899, 'learning_rate/action_model': 6.12903744664413e-05, 'learning_rate/action_query_proj': 6.12903744664413e-05, 'learning_rate/base': 1.5322593616610325e-05, 'epoch': 0.38}) +[2026-06-29 23:28:24,735][starVLA.training.train_starvla][INFO] - Step 877, Loss: {'train/loss': 0.0011738145258277655, 'train/grad_norm_pre_clip': 2.8886449337005615, 'timing/data': 0.00019801408052444458, 'timing/model': 1.2313096669968218, 'learning_rate/action_model': 6.121639055303297e-05, 'learning_rate/action_query_proj': 6.121639055303297e-05, 'learning_rate/base': 1.530409763825824e-05, 'epoch': 0.38}) +[2026-06-29 23:28:57,399][starVLA.training.train_starvla][INFO] - Step 878, Loss: {'train/loss': 0.00021456614194903523, 'train/grad_norm_pre_clip': 1.278724193572998, 'timing/data': 0.00020235194824635983, 'timing/model': 1.2315898379310966, 'learning_rate/action_model': 6.114238389909713e-05, 'learning_rate/action_query_proj': 6.114238389909713e-05, 'learning_rate/base': 1.5285595974774282e-05, 'epoch': 0.38}) +[2026-06-29 23:29:30,038][starVLA.training.train_starvla][INFO] - Step 879, Loss: {'train/loss': 0.011016246862709522, 'train/grad_norm_pre_clip': 1.2460236549377441, 'timing/data': 0.00035361689515411854, 'timing/model': 1.2338081870693713, 'learning_rate/action_model': 6.106835468723782e-05, 'learning_rate/action_query_proj': 6.106835468723782e-05, 'learning_rate/base': 1.5267088671809456e-05, 'epoch': 0.38}) +[2026-06-29 23:30:02,781][starVLA.training.train_starvla][INFO] - Step 880, Loss: {'train/loss': 0.014640918001532555, 'train/grad_norm_pre_clip': 2.0905394554138184, 'timing/data': 0.0002129219938069582, 'timing/model': 1.2428596068639308, 'learning_rate/action_model': 6.099430310011479e-05, 'learning_rate/action_query_proj': 6.099430310011479e-05, 'learning_rate/base': 1.5248575775028697e-05, 'epoch': 0.38}) +[2026-06-29 23:30:35,482][starVLA.training.train_starvla][INFO] - Step 881, Loss: {'train/loss': 5.167465496924706e-05, 'train/grad_norm_pre_clip': 2.1742172241210938, 'timing/data': 0.0002002979163080454, 'timing/model': 1.233177222078666, 'learning_rate/action_model': 6.0920229320442944e-05, 'learning_rate/action_query_proj': 6.0920229320442944e-05, 'learning_rate/base': 1.5230057330110736e-05, 'epoch': 0.38}) +[2026-06-29 23:31:08,176][starVLA.training.train_starvla][INFO] - Step 882, Loss: {'train/loss': 0.03873696178197861, 'train/grad_norm_pre_clip': 2.815436601638794, 'timing/data': 0.00022501498460769653, 'timing/model': 1.2338588410057127, 'learning_rate/action_model': 6.084613353099201e-05, 'learning_rate/action_query_proj': 6.084613353099201e-05, 'learning_rate/base': 1.5211533382748002e-05, 'epoch': 0.38}) +[2026-06-29 23:31:40,857][starVLA.training.train_starvla][INFO] - Step 883, Loss: {'train/loss': 0.000752216496039182, 'train/grad_norm_pre_clip': 4.237499237060547, 'timing/data': 0.00037058000452816486, 'timing/model': 1.2322664689272642, 'learning_rate/action_model': 6.0772015914585933e-05, 'learning_rate/action_query_proj': 6.0772015914585933e-05, 'learning_rate/base': 1.5193003978646483e-05, 'epoch': 0.38}) +[2026-06-29 23:32:13,509][starVLA.training.train_starvla][INFO] - Step 884, Loss: {'train/loss': 0.12187689542770386, 'train/grad_norm_pre_clip': 2.1927473545074463, 'timing/data': 0.00019094999879598618, 'timing/model': 1.2186787449754775, 'learning_rate/action_model': 6.0697876654102594e-05, 'learning_rate/action_query_proj': 6.0697876654102594e-05, 'learning_rate/base': 1.5174469163525649e-05, 'epoch': 0.38}) +[2026-06-29 23:32:46,359][starVLA.training.train_starvla][INFO] - Step 885, Loss: {'train/loss': 0.11867151409387589, 'train/grad_norm_pre_clip': 2.6064798831939697, 'timing/data': 0.00020107999444007874, 'timing/model': 1.2269235129933804, 'learning_rate/action_model': 6.062371593247326e-05, 'learning_rate/action_query_proj': 6.062371593247326e-05, 'learning_rate/base': 1.5155928983118314e-05, 'epoch': 0.38}) +[2026-06-29 23:33:19,035][starVLA.training.train_starvla][INFO] - Step 886, Loss: {'train/loss': 0.026572462171316147, 'train/grad_norm_pre_clip': 2.3085691928863525, 'timing/data': 0.00025542196817696095, 'timing/model': 1.2308620961848646, 'learning_rate/action_model': 6.054953393268209e-05, 'learning_rate/action_query_proj': 6.054953393268209e-05, 'learning_rate/base': 1.5137383483170523e-05, 'epoch': 0.38}) +[2026-06-29 23:33:51,707][starVLA.training.train_starvla][INFO] - Step 887, Loss: {'train/loss': 0.004777414724230766, 'train/grad_norm_pre_clip': 1.96821129322052, 'timing/data': 0.00019718194380402565, 'timing/model': 1.2301171920262277, 'learning_rate/action_model': 6.047533083776584e-05, 'learning_rate/action_query_proj': 6.047533083776584e-05, 'learning_rate/base': 1.511883270944146e-05, 'epoch': 0.38}) +[2026-06-29 23:34:24,405][starVLA.training.train_starvla][INFO] - Step 888, Loss: {'train/loss': 0.443021297454834, 'train/grad_norm_pre_clip': 2.2424957752227783, 'timing/data': 0.00025904783979058266, 'timing/model': 1.2308694529347122, 'learning_rate/action_model': 6.040110683081326e-05, 'learning_rate/action_query_proj': 6.040110683081326e-05, 'learning_rate/base': 1.5100276707703315e-05, 'epoch': 0.38}) +[2026-06-29 23:34:57,069][starVLA.training.train_starvla][INFO] - Step 889, Loss: {'train/loss': 0.4997956156730652, 'train/grad_norm_pre_clip': 2.148376226425171, 'timing/data': 0.00033953110687434673, 'timing/model': 1.2348268320783973, 'learning_rate/action_model': 6.0326862094964695e-05, 'learning_rate/action_query_proj': 6.0326862094964695e-05, 'learning_rate/base': 1.5081715523741174e-05, 'epoch': 0.38}) +[2026-06-29 23:35:29,785][starVLA.training.train_starvla][INFO] - Step 890, Loss: {'train/loss': 0.0006495126872323453, 'train/grad_norm_pre_clip': 2.7628159523010254, 'timing/data': 0.00019863503985106945, 'timing/model': 1.2293168369214982, 'learning_rate/action_model': 6.025259681341166e-05, 'learning_rate/action_query_proj': 6.025259681341166e-05, 'learning_rate/base': 1.5063149203352914e-05, 'epoch': 0.38}) +[2026-06-29 23:36:02,491][starVLA.training.train_starvla][INFO] - Step 891, Loss: {'train/loss': 1.1950656698900275e-05, 'train/grad_norm_pre_clip': 2.296189069747925, 'timing/data': 0.00020178081467747688, 'timing/model': 1.2361944748554379, 'learning_rate/action_model': 6.017831116939636e-05, 'learning_rate/action_query_proj': 6.017831116939636e-05, 'learning_rate/base': 1.504457779234909e-05, 'epoch': 0.38}) +[2026-06-29 23:36:35,203][starVLA.training.train_starvla][INFO] - Step 892, Loss: {'train/loss': 0.0100204823538661, 'train/grad_norm_pre_clip': 1.9495770931243896, 'timing/data': 0.00019863410852849483, 'timing/model': 1.232238742057234, 'learning_rate/action_model': 6.0104005346211225e-05, 'learning_rate/action_query_proj': 6.0104005346211225e-05, 'learning_rate/base': 1.5026001336552806e-05, 'epoch': 0.38}) +[2026-06-29 23:37:07,893][starVLA.training.train_starvla][INFO] - Step 893, Loss: {'train/loss': 2.384104664088227e-05, 'train/grad_norm_pre_clip': 2.6603944301605225, 'timing/data': 0.0004257929977029562, 'timing/model': 1.2285202380735427, 'learning_rate/action_model': 6.00296795271985e-05, 'learning_rate/action_query_proj': 6.00296795271985e-05, 'learning_rate/base': 1.5007419881799625e-05, 'epoch': 0.38}) +[2026-06-29 23:37:40,592][starVLA.training.train_starvla][INFO] - Step 894, Loss: {'train/loss': 0.06696538627147675, 'train/grad_norm_pre_clip': 2.0536508560180664, 'timing/data': 0.000248027965426445, 'timing/model': 1.2335971579886973, 'learning_rate/action_model': 5.995533389574976e-05, 'learning_rate/action_query_proj': 5.995533389574976e-05, 'learning_rate/base': 1.498883347393744e-05, 'epoch': 0.38}) +[2026-06-29 23:38:13,250][starVLA.training.train_starvla][INFO] - Step 895, Loss: {'train/loss': 0.031116915866732597, 'train/grad_norm_pre_clip': 1.5759233236312866, 'timing/data': 0.0003413241356611252, 'timing/model': 1.2319857189431787, 'learning_rate/action_model': 5.9880968635305424e-05, 'learning_rate/action_query_proj': 5.9880968635305424e-05, 'learning_rate/base': 1.4970242158826356e-05, 'epoch': 0.39}) +[2026-06-29 23:38:45,979][starVLA.training.train_starvla][INFO] - Step 896, Loss: {'train/loss': 0.3127378821372986, 'train/grad_norm_pre_clip': 1.6906522512435913, 'timing/data': 0.00020216102711856365, 'timing/model': 1.2305200442206115, 'learning_rate/action_model': 5.9806583929354424e-05, 'learning_rate/action_query_proj': 5.9806583929354424e-05, 'learning_rate/base': 1.4951645982338606e-05, 'epoch': 0.39}) +[2026-06-29 23:39:18,683][starVLA.training.train_starvla][INFO] - Step 897, Loss: {'train/loss': 0.0051270234398543835, 'train/grad_norm_pre_clip': 2.449995994567871, 'timing/data': 0.00019462709315121174, 'timing/model': 1.2302934280596673, 'learning_rate/action_model': 5.9732179961433596e-05, 'learning_rate/action_query_proj': 5.9732179961433596e-05, 'learning_rate/base': 1.4933044990358399e-05, 'epoch': 0.39}) +[2026-06-29 23:39:51,407][starVLA.training.train_starvla][INFO] - Step 898, Loss: {'train/loss': 0.0004739793366752565, 'train/grad_norm_pre_clip': 1.5171877145767212, 'timing/data': 0.00024352897889912128, 'timing/model': 1.240443830145523, 'learning_rate/action_model': 5.965775691512735e-05, 'learning_rate/action_query_proj': 5.965775691512735e-05, 'learning_rate/base': 1.4914439228781838e-05, 'epoch': 0.39}) +[2026-06-29 23:40:24,136][starVLA.training.train_starvla][INFO] - Step 899, Loss: {'train/loss': 0.039947040379047394, 'train/grad_norm_pre_clip': 3.7358429431915283, 'timing/data': 0.00019575003534555435, 'timing/model': 1.2370412310119718, 'learning_rate/action_model': 5.958331497406715e-05, 'learning_rate/action_query_proj': 5.958331497406715e-05, 'learning_rate/base': 1.4895828743516787e-05, 'epoch': 0.39}) +[2026-06-29 23:40:56,821][starVLA.training.train_starvla][INFO] - Step 900, Loss: {'train/loss': 0.1751478761434555, 'train/grad_norm_pre_clip': 1.9864723682403564, 'timing/data': 0.00028942618519067764, 'timing/model': 1.2305473198648542, 'learning_rate/action_model': 5.950885432193108e-05, 'learning_rate/action_query_proj': 5.950885432193108e-05, 'learning_rate/base': 1.487721358048277e-05, 'epoch': 0.39}) +[2026-06-29 23:41:29,475][starVLA.training.train_starvla][INFO] - Step 901, Loss: {'train/loss': 0.00010941080836346373, 'train/grad_norm_pre_clip': 2.842315673828125, 'timing/data': 0.0002473369240760803, 'timing/model': 1.2331463238224387, 'learning_rate/action_model': 5.943437514244341e-05, 'learning_rate/action_query_proj': 5.943437514244341e-05, 'learning_rate/base': 1.4858593785610852e-05, 'epoch': 0.39}) +[2026-06-29 23:42:02,154][starVLA.training.train_starvla][INFO] - Step 902, Loss: {'train/loss': 4.467010876396671e-05, 'train/grad_norm_pre_clip': 2.0323307514190674, 'timing/data': 0.00020656012929975986, 'timing/model': 1.2319989800453186, 'learning_rate/action_model': 5.935987761937412e-05, 'learning_rate/action_query_proj': 5.935987761937412e-05, 'learning_rate/base': 1.483996940484353e-05, 'epoch': 0.39}) +[2026-06-29 23:42:34,833][starVLA.training.train_starvla][INFO] - Step 903, Loss: {'train/loss': 0.006306924857199192, 'train/grad_norm_pre_clip': 3.3313040733337402, 'timing/data': 0.00020100013352930546, 'timing/model': 1.2290276540443301, 'learning_rate/action_model': 5.928536193653841e-05, 'learning_rate/action_query_proj': 5.928536193653841e-05, 'learning_rate/base': 1.4821340484134603e-05, 'epoch': 0.39}) +[2026-06-29 23:43:07,478][starVLA.training.train_starvla][INFO] - Step 904, Loss: {'train/loss': 0.17810429632663727, 'train/grad_norm_pre_clip': 3.072718620300293, 'timing/data': 0.00020073913037776947, 'timing/model': 1.2273214659653604, 'learning_rate/action_model': 5.921082827779635e-05, 'learning_rate/action_query_proj': 5.921082827779635e-05, 'learning_rate/base': 1.4802707069449087e-05, 'epoch': 0.39}) +[2026-06-29 23:43:40,145][starVLA.training.train_starvla][INFO] - Step 905, Loss: {'train/loss': 0.394311785697937, 'train/grad_norm_pre_clip': 3.0424540042877197, 'timing/data': 0.00019745202735066414, 'timing/model': 1.2298078979365528, 'learning_rate/action_model': 5.913627682705232e-05, 'learning_rate/action_query_proj': 5.913627682705232e-05, 'learning_rate/base': 1.478406920676308e-05, 'epoch': 0.39}) +[2026-06-29 23:44:12,844][starVLA.training.train_starvla][INFO] - Step 906, Loss: {'train/loss': 0.01192055456340313, 'train/grad_norm_pre_clip': 1.74856436252594, 'timing/data': 0.0001939958892762661, 'timing/model': 1.234294984024018, 'learning_rate/action_model': 5.906170776825459e-05, 'learning_rate/action_query_proj': 5.906170776825459e-05, 'learning_rate/base': 1.4765426942063648e-05, 'epoch': 0.39}) +[2026-06-29 23:44:45,537][starVLA.training.train_starvla][INFO] - Step 907, Loss: {'train/loss': 0.03191310912370682, 'train/grad_norm_pre_clip': 3.7451558113098145, 'timing/data': 0.00019542896188795567, 'timing/model': 1.2318235421553254, 'learning_rate/action_model': 5.8987121285394965e-05, 'learning_rate/action_query_proj': 5.8987121285394965e-05, 'learning_rate/base': 1.4746780321348741e-05, 'epoch': 0.39}) +[2026-06-29 23:45:18,244][starVLA.training.train_starvla][INFO] - Step 908, Loss: {'train/loss': 1.3709058066524449e-06, 'train/grad_norm_pre_clip': 4.373441219329834, 'timing/data': 0.00030651804991066456, 'timing/model': 1.2306722609791905, 'learning_rate/action_model': 5.8912517562508104e-05, 'learning_rate/action_query_proj': 5.8912517562508104e-05, 'learning_rate/base': 1.4728129390627026e-05, 'epoch': 0.39}) +[2026-06-29 23:45:50,952][starVLA.training.train_starvla][INFO] - Step 909, Loss: {'train/loss': 0.00010751008812803775, 'train/grad_norm_pre_clip': 4.584482669830322, 'timing/data': 0.00020872289314866066, 'timing/model': 1.2316562568303198, 'learning_rate/action_model': 5.883789678367134e-05, 'learning_rate/action_query_proj': 5.883789678367134e-05, 'learning_rate/base': 1.4709474195917836e-05, 'epoch': 0.39}) +[2026-06-29 23:46:23,658][starVLA.training.train_starvla][INFO] - Step 910, Loss: {'train/loss': 0.001448804046958685, 'train/grad_norm_pre_clip': 4.713263988494873, 'timing/data': 0.00020343391224741936, 'timing/model': 1.23638976784423, 'learning_rate/action_model': 5.876325913300397e-05, 'learning_rate/action_query_proj': 5.876325913300397e-05, 'learning_rate/base': 1.4690814783250992e-05, 'epoch': 0.39}) +[2026-06-29 23:46:56,369][starVLA.training.train_starvla][INFO] - Step 911, Loss: {'train/loss': 0.5911746621131897, 'train/grad_norm_pre_clip': 1.7857749462127686, 'timing/data': 0.00019738194532692432, 'timing/model': 1.231503410032019, 'learning_rate/action_model': 5.868860479466702e-05, 'learning_rate/action_query_proj': 5.868860479466702e-05, 'learning_rate/base': 1.4672151198666756e-05, 'epoch': 0.39}) +[2026-06-29 23:47:29,082][starVLA.training.train_starvla][INFO] - Step 912, Loss: {'train/loss': 0.1374107003211975, 'train/grad_norm_pre_clip': 2.1266322135925293, 'timing/data': 0.00019295397214591503, 'timing/model': 1.2316256209742278, 'learning_rate/action_model': 5.8613933952862634e-05, 'learning_rate/action_query_proj': 5.8613933952862634e-05, 'learning_rate/base': 1.4653483488215659e-05, 'epoch': 0.39}) +[2026-06-29 23:48:01,764][starVLA.training.train_starvla][INFO] - Step 913, Loss: {'train/loss': 0.029575757682323456, 'train/grad_norm_pre_clip': 1.6955164670944214, 'timing/data': 0.0004952249582856894, 'timing/model': 1.230507472064346, 'learning_rate/action_model': 5.8539246791833676e-05, 'learning_rate/action_query_proj': 5.8539246791833676e-05, 'learning_rate/base': 1.4634811697958419e-05, 'epoch': 0.39}) +[2026-06-29 23:48:34,431][starVLA.training.train_starvla][INFO] - Step 914, Loss: {'train/loss': 0.0019517616601660848, 'train/grad_norm_pre_clip': 3.82194185256958, 'timing/data': 0.0002364960964769125, 'timing/model': 1.234565587947145, 'learning_rate/action_model': 5.846454349586333e-05, 'learning_rate/action_query_proj': 5.846454349586333e-05, 'learning_rate/base': 1.4616135873965833e-05, 'epoch': 0.39}) +[2026-06-29 23:49:07,107][starVLA.training.train_starvla][INFO] - Step 915, Loss: {'train/loss': 0.18944492936134338, 'train/grad_norm_pre_clip': 3.6324870586395264, 'timing/data': 0.0002711820416152477, 'timing/model': 1.234962227055803, 'learning_rate/action_model': 5.838982424927452e-05, 'learning_rate/action_query_proj': 5.838982424927452e-05, 'learning_rate/base': 1.459745606231863e-05, 'epoch': 0.39}) +[2026-06-29 23:49:39,823][starVLA.training.train_starvla][INFO] - Step 916, Loss: {'train/loss': 0.0248394962400198, 'train/grad_norm_pre_clip': 2.4669244289398193, 'timing/data': 0.0002680548932403326, 'timing/model': 1.2299899710342288, 'learning_rate/action_model': 5.8315089236429566e-05, 'learning_rate/action_query_proj': 5.8315089236429566e-05, 'learning_rate/base': 1.4578772309107391e-05, 'epoch': 0.39}) +[2026-06-29 23:50:12,549][starVLA.training.train_starvla][INFO] - Step 917, Loss: {'train/loss': 0.00818036962300539, 'train/grad_norm_pre_clip': 1.518843412399292, 'timing/data': 0.00019686203449964523, 'timing/model': 1.231975540984422, 'learning_rate/action_model': 5.824033864172969e-05, 'learning_rate/action_query_proj': 5.824033864172969e-05, 'learning_rate/base': 1.4560084660432422e-05, 'epoch': 0.39}) +[2026-06-29 23:50:45,244][starVLA.training.train_starvla][INFO] - Step 918, Loss: {'train/loss': 0.012564277276396751, 'train/grad_norm_pre_clip': 2.1908411979675293, 'timing/data': 0.00022909301333129406, 'timing/model': 1.232634006999433, 'learning_rate/action_model': 5.816557264961454e-05, 'learning_rate/action_query_proj': 5.816557264961454e-05, 'learning_rate/base': 1.4541393162403635e-05, 'epoch': 0.39}) +[2026-06-29 23:51:17,914][starVLA.training.train_starvla][INFO] - Step 919, Loss: {'train/loss': 0.06004157289862633, 'train/grad_norm_pre_clip': 6.165485858917236, 'timing/data': 0.0002562841400504112, 'timing/model': 1.2291814170312136, 'learning_rate/action_model': 5.809079144456178e-05, 'learning_rate/action_query_proj': 5.809079144456178e-05, 'learning_rate/base': 1.4522697861140444e-05, 'epoch': 0.4}) +[2026-06-29 23:51:50,602][starVLA.training.train_starvla][INFO] - Step 920, Loss: {'train/loss': 0.0001300357689615339, 'train/grad_norm_pre_clip': 1.4985792636871338, 'timing/data': 0.00029378500767052174, 'timing/model': 1.2305402650963515, 'learning_rate/action_model': 5.80159952110866e-05, 'learning_rate/action_query_proj': 5.80159952110866e-05, 'learning_rate/base': 1.450399880277165e-05, 'epoch': 0.4}) +[2026-06-29 23:52:23,286][starVLA.training.train_starvla][INFO] - Step 921, Loss: {'train/loss': 0.0006055932608433068, 'train/grad_norm_pre_clip': 1.9134953022003174, 'timing/data': 0.00019755191169679165, 'timing/model': 1.23085947195068, 'learning_rate/action_model': 5.7941184133741274e-05, 'learning_rate/action_query_proj': 5.7941184133741274e-05, 'learning_rate/base': 1.4485296033435319e-05, 'epoch': 0.4}) +[2026-06-29 23:52:55,981][starVLA.training.train_starvla][INFO] - Step 922, Loss: {'train/loss': 0.05016160011291504, 'train/grad_norm_pre_clip': 1.4940714836120605, 'timing/data': 0.00020094914361834526, 'timing/model': 1.2325890602078289, 'learning_rate/action_model': 5.78663583971147e-05, 'learning_rate/action_query_proj': 5.78663583971147e-05, 'learning_rate/base': 1.4466589599278674e-05, 'epoch': 0.4}) +[2026-06-29 23:53:28,697][starVLA.training.train_starvla][INFO] - Step 923, Loss: {'train/loss': 3.3050033380277455e-05, 'train/grad_norm_pre_clip': 2.1160378456115723, 'timing/data': 0.00031024497002363205, 'timing/model': 1.2301778870169073, 'learning_rate/action_model': 5.779151818583194e-05, 'learning_rate/action_query_proj': 5.779151818583194e-05, 'learning_rate/base': 1.4447879546457985e-05, 'epoch': 0.4}) +[2026-06-29 23:54:01,436][starVLA.training.train_starvla][INFO] - Step 924, Loss: {'train/loss': 0.0009648798150010407, 'train/grad_norm_pre_clip': 2.198396682739258, 'timing/data': 0.00020329398103058338, 'timing/model': 1.2369621430989355, 'learning_rate/action_model': 5.7716663684553754e-05, 'learning_rate/action_query_proj': 5.7716663684553754e-05, 'learning_rate/base': 1.4429165921138438e-05, 'epoch': 0.4}) +[2026-06-29 23:54:34,138][starVLA.training.train_starvla][INFO] - Step 925, Loss: {'train/loss': 0.005485032685101032, 'train/grad_norm_pre_clip': 1.6892788410186768, 'timing/data': 0.00027097086422145367, 'timing/model': 1.234744715038687, 'learning_rate/action_model': 5.7641795077976214e-05, 'learning_rate/action_query_proj': 5.7641795077976214e-05, 'learning_rate/base': 1.4410448769494053e-05, 'epoch': 0.4}) +[2026-06-29 23:55:06,880][starVLA.training.train_starvla][INFO] - Step 926, Loss: {'train/loss': 0.07290603220462799, 'train/grad_norm_pre_clip': 2.779550552368164, 'timing/data': 0.00022577610798180103, 'timing/model': 1.228434334974736, 'learning_rate/action_model': 5.756691255083015e-05, 'learning_rate/action_query_proj': 5.756691255083015e-05, 'learning_rate/base': 1.4391728137707538e-05, 'epoch': 0.4}) +[2026-06-29 23:55:39,597][starVLA.training.train_starvla][INFO] - Step 927, Loss: {'train/loss': 0.17395223677158356, 'train/grad_norm_pre_clip': 3.8972184658050537, 'timing/data': 0.00020413496531546116, 'timing/model': 1.23265957320109, 'learning_rate/action_model': 5.749201628788077e-05, 'learning_rate/action_query_proj': 5.749201628788077e-05, 'learning_rate/base': 1.4373004071970193e-05, 'epoch': 0.4}) +[2026-06-29 23:56:12,267][starVLA.training.train_starvla][INFO] - Step 928, Loss: {'train/loss': 0.25778913497924805, 'train/grad_norm_pre_clip': 2.675135374069214, 'timing/data': 0.000454107066616416, 'timing/model': 1.2339660751167685, 'learning_rate/action_model': 5.741710647392712e-05, 'learning_rate/action_query_proj': 5.741710647392712e-05, 'learning_rate/base': 1.435427661848178e-05, 'epoch': 0.4}) +[2026-06-29 23:56:44,943][starVLA.training.train_starvla][INFO] - Step 929, Loss: {'train/loss': 0.00048277279711328447, 'train/grad_norm_pre_clip': 2.0908939838409424, 'timing/data': 0.00019860407337546349, 'timing/model': 1.2307748799212277, 'learning_rate/action_model': 5.734218329380177e-05, 'learning_rate/action_query_proj': 5.734218329380177e-05, 'learning_rate/base': 1.4335545823450442e-05, 'epoch': 0.4}) +[2026-06-29 23:57:17,661][starVLA.training.train_starvla][INFO] - Step 930, Loss: {'train/loss': 0.008156487718224525, 'train/grad_norm_pre_clip': 2.8648386001586914, 'timing/data': 0.0002011891920119524, 'timing/model': 1.2359211870934814, 'learning_rate/action_model': 5.726724693237017e-05, 'learning_rate/action_query_proj': 5.726724693237017e-05, 'learning_rate/base': 1.4316811733092542e-05, 'epoch': 0.4}) +[2026-06-29 23:57:50,355][starVLA.training.train_starvla][INFO] - Step 931, Loss: {'train/loss': 0.03078570030629635, 'train/grad_norm_pre_clip': 2.753589391708374, 'timing/data': 0.00020103994756937027, 'timing/model': 1.2361290270928293, 'learning_rate/action_model': 5.7192297574530396e-05, 'learning_rate/action_query_proj': 5.7192297574530396e-05, 'learning_rate/base': 1.4298074393632599e-05, 'epoch': 0.4}) +[2026-06-29 23:58:23,022][starVLA.training.train_starvla][INFO] - Step 932, Loss: {'train/loss': 0.21953828632831573, 'train/grad_norm_pre_clip': 1.5871561765670776, 'timing/data': 0.00032785884104669094, 'timing/model': 1.2271113300230354, 'learning_rate/action_model': 5.711733540521248e-05, 'learning_rate/action_query_proj': 5.711733540521248e-05, 'learning_rate/base': 1.427933385130312e-05, 'epoch': 0.4}) +[2026-06-29 23:58:55,674][starVLA.training.train_starvla][INFO] - Step 933, Loss: {'train/loss': 0.4921084940433502, 'train/grad_norm_pre_clip': 1.7604849338531494, 'timing/data': 0.00021560583263635635, 'timing/model': 1.2304096810985357, 'learning_rate/action_model': 5.704236060937817e-05, 'learning_rate/action_query_proj': 5.704236060937817e-05, 'learning_rate/base': 1.4260590152344542e-05, 'epoch': 0.4}) +[2026-06-29 23:59:28,334][starVLA.training.train_starvla][INFO] - Step 934, Loss: {'train/loss': 0.02073306031525135, 'train/grad_norm_pre_clip': 2.0919134616851807, 'timing/data': 0.0001915509346872568, 'timing/model': 1.228470369009301, 'learning_rate/action_model': 5.6967373372020285e-05, 'learning_rate/action_query_proj': 5.6967373372020285e-05, 'learning_rate/base': 1.4241843343005071e-05, 'epoch': 0.4}) +[2026-06-30 00:00:01,008][starVLA.training.train_starvla][INFO] - Step 935, Loss: {'train/loss': 0.0011270770337432623, 'train/grad_norm_pre_clip': 1.7753386497497559, 'timing/data': 0.00019601895473897457, 'timing/model': 1.2333748508244753, 'learning_rate/action_model': 5.689237387816241e-05, 'learning_rate/action_query_proj': 5.689237387816241e-05, 'learning_rate/base': 1.4223093469540602e-05, 'epoch': 0.4}) +[2026-06-30 00:00:33,698][starVLA.training.train_starvla][INFO] - Step 936, Loss: {'train/loss': 0.13972724974155426, 'train/grad_norm_pre_clip': 3.023193120956421, 'timing/data': 0.00026937806978821754, 'timing/model': 1.2337456189561635, 'learning_rate/action_model': 5.681736231285831e-05, 'learning_rate/action_query_proj': 5.681736231285831e-05, 'learning_rate/base': 1.4204340578214578e-05, 'epoch': 0.4}) +[2026-06-30 00:01:06,337][starVLA.training.train_starvla][INFO] - Step 937, Loss: {'train/loss': 0.22157524526119232, 'train/grad_norm_pre_clip': 1.5436577796936035, 'timing/data': 0.00030020601116120815, 'timing/model': 1.2365288969594985, 'learning_rate/action_model': 5.67423388611916e-05, 'learning_rate/action_query_proj': 5.67423388611916e-05, 'learning_rate/base': 1.41855847152979e-05, 'epoch': 0.4}) +[2026-06-30 00:01:39,242][starVLA.training.train_starvla][INFO] - Step 938, Loss: {'train/loss': 0.014097531326115131, 'train/grad_norm_pre_clip': 1.4549931287765503, 'timing/data': 0.00029865303076803684, 'timing/model': 1.2300237820018083, 'learning_rate/action_model': 5.666730370827519e-05, 'learning_rate/action_query_proj': 5.666730370827519e-05, 'learning_rate/base': 1.4166825927068797e-05, 'epoch': 0.4}) +[2026-06-30 00:02:11,907][starVLA.training.train_starvla][INFO] - Step 939, Loss: {'train/loss': 8.31967918202281e-05, 'train/grad_norm_pre_clip': 1.9872760772705078, 'timing/data': 0.0002014299388974905, 'timing/model': 1.2268047481775284, 'learning_rate/action_model': 5.659225703925081e-05, 'learning_rate/action_query_proj': 5.659225703925081e-05, 'learning_rate/base': 1.4148064259812702e-05, 'epoch': 0.4}) +[2026-06-30 00:02:44,540][starVLA.training.train_starvla][INFO] - Step 940, Loss: {'train/loss': 0.014775291085243225, 'train/grad_norm_pre_clip': 2.3443820476531982, 'timing/data': 0.00019428692758083344, 'timing/model': 1.2340465399902314, 'learning_rate/action_model': 5.651719903928871e-05, 'learning_rate/action_query_proj': 5.651719903928871e-05, 'learning_rate/base': 1.4129299759822177e-05, 'epoch': 0.4}) +[2026-06-30 00:03:17,202][starVLA.training.train_starvla][INFO] - Step 941, Loss: {'train/loss': 1.656994572840631e-05, 'train/grad_norm_pre_clip': 2.397554397583008, 'timing/data': 0.0001901579089462757, 'timing/model': 1.2345970761962235, 'learning_rate/action_model': 5.6442129893587e-05, 'learning_rate/action_query_proj': 5.6442129893587e-05, 'learning_rate/base': 1.411053247339675e-05, 'epoch': 0.4}) +[2026-06-30 00:03:49,896][starVLA.training.train_starvla][INFO] - Step 942, Loss: {'train/loss': 0.00015685682592447847, 'train/grad_norm_pre_clip': 2.902249813079834, 'timing/data': 0.0002014001365751028, 'timing/model': 1.2328990511596203, 'learning_rate/action_model': 5.6367049787371373e-05, 'learning_rate/action_query_proj': 5.6367049787371373e-05, 'learning_rate/base': 1.4091762446842843e-05, 'epoch': 0.41}) +[2026-06-30 00:04:22,634][starVLA.training.train_starvla][INFO] - Step 943, Loss: {'train/loss': 0.29295802116394043, 'train/grad_norm_pre_clip': 2.2607133388519287, 'timing/data': 0.00020001805387437344, 'timing/model': 1.2332667040172964, 'learning_rate/action_model': 5.6291958905894494e-05, 'learning_rate/action_query_proj': 5.6291958905894494e-05, 'learning_rate/base': 1.4072989726473624e-05, 'epoch': 0.41}) +[2026-06-30 00:04:55,320][starVLA.training.train_starvla][INFO] - Step 944, Loss: {'train/loss': 0.3086633086204529, 'train/grad_norm_pre_clip': 2.98885440826416, 'timing/data': 0.000249149976298213, 'timing/model': 1.2230335650965571, 'learning_rate/action_model': 5.6216857434435687e-05, 'learning_rate/action_query_proj': 5.6216857434435687e-05, 'learning_rate/base': 1.4054214358608922e-05, 'epoch': 0.41}) +[2026-06-30 00:05:27,976][starVLA.training.train_starvla][INFO] - Step 945, Loss: {'train/loss': 0.0005888580344617367, 'train/grad_norm_pre_clip': 2.971329927444458, 'timing/data': 0.00019445689395070076, 'timing/model': 1.2313499951269478, 'learning_rate/action_model': 5.614174555830034e-05, 'learning_rate/action_query_proj': 5.614174555830034e-05, 'learning_rate/base': 1.4035436389575085e-05, 'epoch': 0.41}) +[2026-06-30 00:06:00,646][starVLA.training.train_starvla][INFO] - Step 946, Loss: {'train/loss': 8.28500105853891e-06, 'train/grad_norm_pre_clip': 1.9416667222976685, 'timing/data': 0.0003145630471408367, 'timing/model': 1.2383045959286392, 'learning_rate/action_model': 5.606662346281953e-05, 'learning_rate/action_query_proj': 5.606662346281953e-05, 'learning_rate/base': 1.4016655865704883e-05, 'epoch': 0.41}) +[2026-06-30 00:06:33,307][starVLA.training.train_starvla][INFO] - Step 947, Loss: {'train/loss': 3.31392147927545e-05, 'train/grad_norm_pre_clip': 2.8225255012512207, 'timing/data': 0.00029112910851836205, 'timing/model': 1.219382253009826, 'learning_rate/action_model': 5.59914913333496e-05, 'learning_rate/action_query_proj': 5.59914913333496e-05, 'learning_rate/base': 1.39978728333374e-05, 'epoch': 0.41}) +[2026-06-30 00:07:05,994][starVLA.training.train_starvla][INFO] - Step 948, Loss: {'train/loss': 0.00012928569049108773, 'train/grad_norm_pre_clip': 2.181626081466675, 'timing/data': 0.0002548608463257551, 'timing/model': 1.228387092007324, 'learning_rate/action_model': 5.591634935527158e-05, 'learning_rate/action_query_proj': 5.591634935527158e-05, 'learning_rate/base': 1.3979087338817894e-05, 'epoch': 0.41}) +[2026-06-30 00:07:38,675][starVLA.training.train_starvla][INFO] - Step 949, Loss: {'train/loss': 0.2513733506202698, 'train/grad_norm_pre_clip': 1.8471754789352417, 'timing/data': 0.00022517493925988674, 'timing/model': 1.2325041389558464, 'learning_rate/action_model': 5.584119771399083e-05, 'learning_rate/action_query_proj': 5.584119771399083e-05, 'learning_rate/base': 1.3960299428497708e-05, 'epoch': 0.41}) +[2026-06-30 00:08:11,356][starVLA.training.train_starvla][INFO] - Step 950, Loss: {'train/loss': 0.06039254367351532, 'train/grad_norm_pre_clip': 1.523526668548584, 'timing/data': 0.00038617895916104317, 'timing/model': 1.2332274930085987, 'learning_rate/action_model': 5.576603659493657e-05, 'learning_rate/action_query_proj': 5.576603659493657e-05, 'learning_rate/base': 1.3941509148734142e-05, 'epoch': 0.41}) +[2026-06-30 00:08:44,010][starVLA.training.train_starvla][INFO] - Step 951, Loss: {'train/loss': 0.19078165292739868, 'train/grad_norm_pre_clip': 0.9614737033843994, 'timing/data': 0.0003581950441002846, 'timing/model': 1.2333853701129556, 'learning_rate/action_model': 5.5690866183561383e-05, 'learning_rate/action_query_proj': 5.5690866183561383e-05, 'learning_rate/base': 1.3922716545890346e-05, 'epoch': 0.41}) +[2026-06-30 00:09:16,688][starVLA.training.train_starvla][INFO] - Step 952, Loss: {'train/loss': 0.23621684312820435, 'train/grad_norm_pre_clip': 1.4426684379577637, 'timing/data': 0.0002359549980610609, 'timing/model': 1.2331685898825526, 'learning_rate/action_model': 5.561568666534078e-05, 'learning_rate/action_query_proj': 5.561568666534078e-05, 'learning_rate/base': 1.3903921666335195e-05, 'epoch': 0.41}) +[2026-06-30 00:09:49,406][starVLA.training.train_starvla][INFO] - Step 953, Loss: {'train/loss': 0.011435262858867645, 'train/grad_norm_pre_clip': 1.993643045425415, 'timing/data': 0.0001916710752993822, 'timing/model': 1.2286596021149307, 'learning_rate/action_model': 5.5540498225772765e-05, 'learning_rate/action_query_proj': 5.5540498225772765e-05, 'learning_rate/base': 1.3885124556443191e-05, 'epoch': 0.41}) +[2026-06-30 00:10:22,070][starVLA.training.train_starvla][INFO] - Step 954, Loss: {'train/loss': 0.033985935151576996, 'train/grad_norm_pre_clip': 1.703901767730713, 'timing/data': 0.00020183110609650612, 'timing/model': 1.2329538939520717, 'learning_rate/action_model': 5.546530105037733e-05, 'learning_rate/action_query_proj': 5.546530105037733e-05, 'learning_rate/base': 1.3866325262594332e-05, 'epoch': 0.41}) +[2026-06-30 00:10:54,751][starVLA.training.train_starvla][INFO] - Step 955, Loss: {'train/loss': 0.0001797063450794667, 'train/grad_norm_pre_clip': 2.3112223148345947, 'timing/data': 0.00026477896608412266, 'timing/model': 1.2303234501741827, 'learning_rate/action_model': 5.539009532469601e-05, 'learning_rate/action_query_proj': 5.539009532469601e-05, 'learning_rate/base': 1.3847523831174002e-05, 'epoch': 0.41}) +[2026-06-30 00:11:27,468][starVLA.training.train_starvla][INFO] - Step 956, Loss: {'train/loss': 0.006797497626394033, 'train/grad_norm_pre_clip': 1.239785075187683, 'timing/data': 0.00020329304970800877, 'timing/model': 1.2362423371523619, 'learning_rate/action_model': 5.531488123429149e-05, 'learning_rate/action_query_proj': 5.531488123429149e-05, 'learning_rate/base': 1.3828720308572873e-05, 'epoch': 0.41}) +[2026-06-30 00:12:00,135][starVLA.training.train_starvla][INFO] - Step 957, Loss: {'train/loss': 0.00017303724598605186, 'train/grad_norm_pre_clip': 2.461483955383301, 'timing/data': 0.0002466151490807533, 'timing/model': 1.2324967698659748, 'learning_rate/action_model': 5.5239658964747045e-05, 'learning_rate/action_query_proj': 5.5239658964747045e-05, 'learning_rate/base': 1.3809914741186761e-05, 'epoch': 0.41}) +[2026-06-30 00:12:32,787][starVLA.training.train_starvla][INFO] - Step 958, Loss: {'train/loss': 0.5362901091575623, 'train/grad_norm_pre_clip': 1.817327857017517, 'timing/data': 0.0002027619630098343, 'timing/model': 1.2345922419335693, 'learning_rate/action_model': 5.516442870166615e-05, 'learning_rate/action_query_proj': 5.516442870166615e-05, 'learning_rate/base': 1.3791107175416537e-05, 'epoch': 0.41}) +[2026-06-30 00:13:05,481][starVLA.training.train_starvla][INFO] - Step 959, Loss: {'train/loss': 0.03915312886238098, 'train/grad_norm_pre_clip': 5.000826358795166, 'timing/data': 0.00019273394718766212, 'timing/model': 1.2290422609075904, 'learning_rate/action_model': 5.5089190630671995e-05, 'learning_rate/action_query_proj': 5.5089190630671995e-05, 'learning_rate/base': 1.3772297657667999e-05, 'epoch': 0.41}) +[2026-06-30 00:13:38,177][starVLA.training.train_starvla][INFO] - Step 960, Loss: {'train/loss': 0.0014082746347412467, 'train/grad_norm_pre_clip': 3.943157911300659, 'timing/data': 0.000294526107609272, 'timing/model': 1.2324840938672423, 'learning_rate/action_model': 5.5013944937407045e-05, 'learning_rate/action_query_proj': 5.5013944937407045e-05, 'learning_rate/base': 1.3753486234351761e-05, 'epoch': 0.41}) +[2026-06-30 00:14:10,895][starVLA.training.train_starvla][INFO] - Step 961, Loss: {'train/loss': 0.4633047580718994, 'train/grad_norm_pre_clip': 2.1012744903564453, 'timing/data': 0.00034477096050977707, 'timing/model': 1.228450994938612, 'learning_rate/action_model': 5.4938691807532574e-05, 'learning_rate/action_query_proj': 5.4938691807532574e-05, 'learning_rate/base': 1.3734672951883143e-05, 'epoch': 0.41}) +[2026-06-30 00:14:43,573][starVLA.training.train_starvla][INFO] - Step 962, Loss: {'train/loss': 0.00010413912241347134, 'train/grad_norm_pre_clip': 2.3417065143585205, 'timing/data': 0.00019239308312535286, 'timing/model': 1.2288147599902004, 'learning_rate/action_model': 5.4863431426728204e-05, 'learning_rate/action_query_proj': 5.4863431426728204e-05, 'learning_rate/base': 1.3715857856682051e-05, 'epoch': 0.41}) +[2026-06-30 00:15:16,235][starVLA.training.train_starvla][INFO] - Step 963, Loss: {'train/loss': 0.6178090572357178, 'train/grad_norm_pre_clip': 1.8683521747589111, 'timing/data': 0.00020282319746911526, 'timing/model': 1.2276157678570598, 'learning_rate/action_model': 5.478816398069143e-05, 'learning_rate/action_query_proj': 5.478816398069143e-05, 'learning_rate/base': 1.3697040995172857e-05, 'epoch': 0.41}) +[2026-06-30 00:15:48,957][starVLA.training.train_starvla][INFO] - Step 964, Loss: {'train/loss': 0.5254952311515808, 'train/grad_norm_pre_clip': 1.8637452125549316, 'timing/data': 0.0002647889778017998, 'timing/model': 1.2317830598913133, 'learning_rate/action_model': 5.471288965513719e-05, 'learning_rate/action_query_proj': 5.471288965513719e-05, 'learning_rate/base': 1.3678222413784298e-05, 'epoch': 0.41}) +[2026-06-30 00:16:21,638][starVLA.training.train_starvla][INFO] - Step 965, Loss: {'train/loss': 0.4992577135562897, 'train/grad_norm_pre_clip': 2.294504165649414, 'timing/data': 0.00023669679649174213, 'timing/model': 1.2273482119198889, 'learning_rate/action_model': 5.463760863579741e-05, 'learning_rate/action_query_proj': 5.463760863579741e-05, 'learning_rate/base': 1.3659402158949353e-05, 'epoch': 0.42}) +[2026-06-30 00:16:54,327][starVLA.training.train_starvla][INFO] - Step 966, Loss: {'train/loss': 0.42985811829566956, 'train/grad_norm_pre_clip': 3.1023693084716797, 'timing/data': 0.00024345004931092262, 'timing/model': 1.2300859680399299, 'learning_rate/action_model': 5.456232110842051e-05, 'learning_rate/action_query_proj': 5.456232110842051e-05, 'learning_rate/base': 1.3640580277105128e-05, 'epoch': 0.42}) +[2026-06-30 00:17:26,994][starVLA.training.train_starvla][INFO] - Step 967, Loss: {'train/loss': 0.49279549717903137, 'train/grad_norm_pre_clip': 2.676100015640259, 'timing/data': 0.00019862502813339233, 'timing/model': 1.234357938170433, 'learning_rate/action_model': 5.448702725877101e-05, 'learning_rate/action_query_proj': 5.448702725877101e-05, 'learning_rate/base': 1.3621756814692752e-05, 'epoch': 0.42}) +[2026-06-30 00:17:59,690][starVLA.training.train_starvla][INFO] - Step 968, Loss: {'train/loss': 0.04061011224985123, 'train/grad_norm_pre_clip': 1.5708528757095337, 'timing/data': 0.0002984441816806793, 'timing/model': 1.2296526690479368, 'learning_rate/action_model': 5.4411727272628945e-05, 'learning_rate/action_query_proj': 5.4411727272628945e-05, 'learning_rate/base': 1.3602931818157236e-05, 'epoch': 0.42}) +[2026-06-30 00:18:32,364][starVLA.training.train_starvla][INFO] - Step 969, Loss: {'train/loss': 0.3907739222049713, 'train/grad_norm_pre_clip': 3.1775670051574707, 'timing/data': 0.0002535278908908367, 'timing/model': 1.2299494189210236, 'learning_rate/action_model': 5.43364213357896e-05, 'learning_rate/action_query_proj': 5.43364213357896e-05, 'learning_rate/base': 1.35841053339474e-05, 'epoch': 0.42}) +[2026-06-30 00:19:05,062][starVLA.training.train_starvla][INFO] - Step 970, Loss: {'train/loss': 0.008048666641116142, 'train/grad_norm_pre_clip': 2.375680923461914, 'timing/data': 0.00019720196723937988, 'timing/model': 1.2348026661202312, 'learning_rate/action_model': 5.426110963406284e-05, 'learning_rate/action_query_proj': 5.426110963406284e-05, 'learning_rate/base': 1.356527740851571e-05, 'epoch': 0.42}) +[2026-06-30 00:19:37,727][starVLA.training.train_starvla][INFO] - Step 971, Loss: {'train/loss': 0.31410497426986694, 'train/grad_norm_pre_clip': 2.054933786392212, 'timing/data': 0.0001888058613985777, 'timing/model': 1.232347572920844, 'learning_rate/action_model': 5.418579235327287e-05, 'learning_rate/action_query_proj': 5.418579235327287e-05, 'learning_rate/base': 1.3546448088318217e-05, 'epoch': 0.42}) +[2026-06-30 00:20:10,444][starVLA.training.train_starvla][INFO] - Step 972, Loss: {'train/loss': 0.06419935822486877, 'train/grad_norm_pre_clip': 1.7846146821975708, 'timing/data': 0.0002961978316307068, 'timing/model': 1.232488548848778, 'learning_rate/action_model': 5.411046967925751e-05, 'learning_rate/action_query_proj': 5.411046967925751e-05, 'learning_rate/base': 1.3527617419814378e-05, 'epoch': 0.42}) +[2026-06-30 00:20:43,127][starVLA.training.train_starvla][INFO] - Step 973, Loss: {'train/loss': 0.46463820338249207, 'train/grad_norm_pre_clip': 1.5528610944747925, 'timing/data': 0.0003058570437133312, 'timing/model': 1.233504154952243, 'learning_rate/action_model': 5.403514179786805e-05, 'learning_rate/action_query_proj': 5.403514179786805e-05, 'learning_rate/base': 1.3508785449467013e-05, 'epoch': 0.42}) +[2026-06-30 00:21:15,827][starVLA.training.train_starvla][INFO] - Step 974, Loss: {'train/loss': 0.21068334579467773, 'train/grad_norm_pre_clip': 2.140242099761963, 'timing/data': 0.00037104985676705837, 'timing/model': 1.2322729320731014, 'learning_rate/action_model': 5.3959808894968486e-05, 'learning_rate/action_query_proj': 5.3959808894968486e-05, 'learning_rate/base': 1.3489952223742122e-05, 'epoch': 0.42}) +[2026-06-30 00:21:48,491][starVLA.training.train_starvla][INFO] - Step 975, Loss: {'train/loss': 0.000388596294214949, 'train/grad_norm_pre_clip': 1.9925085306167603, 'timing/data': 0.00019520800560712814, 'timing/model': 1.229777042986825, 'learning_rate/action_model': 5.38844711564353e-05, 'learning_rate/action_query_proj': 5.38844711564353e-05, 'learning_rate/base': 1.3471117789108825e-05, 'epoch': 0.42}) +[2026-06-30 00:22:21,186][starVLA.training.train_starvla][INFO] - Step 976, Loss: {'train/loss': 0.02333996631205082, 'train/grad_norm_pre_clip': 3.0190281867980957, 'timing/data': 0.00022320193238556385, 'timing/model': 1.2321721808984876, 'learning_rate/action_model': 5.380912876815687e-05, 'learning_rate/action_query_proj': 5.380912876815687e-05, 'learning_rate/base': 1.3452282192039217e-05, 'epoch': 0.42}) +[2026-06-30 00:22:53,830][starVLA.training.train_starvla][INFO] - Step 977, Loss: {'train/loss': 0.0005023261765018106, 'train/grad_norm_pre_clip': 1.2239855527877808, 'timing/data': 0.00027555995620787144, 'timing/model': 1.2296236627735198, 'learning_rate/action_model': 5.3733781916033036e-05, 'learning_rate/action_query_proj': 5.3733781916033036e-05, 'learning_rate/base': 1.3433445479008259e-05, 'epoch': 0.42}) +[2026-06-30 00:23:26,519][starVLA.training.train_starvla][INFO] - Step 978, Loss: {'train/loss': 0.08520466834306717, 'train/grad_norm_pre_clip': 1.6850700378417969, 'timing/data': 0.00019087106920778751, 'timing/model': 1.2300325469113886, 'learning_rate/action_model': 5.3658430785974675e-05, 'learning_rate/action_query_proj': 5.3658430785974675e-05, 'learning_rate/base': 1.3414607696493669e-05, 'epoch': 0.42}) +[2026-06-30 00:23:59,191][starVLA.training.train_starvla][INFO] - Step 979, Loss: {'train/loss': 0.0542684867978096, 'train/grad_norm_pre_clip': 1.7805004119873047, 'timing/data': 0.00019858498126268387, 'timing/model': 1.2261390248313546, 'learning_rate/action_model': 5.358307556390321e-05, 'learning_rate/action_query_proj': 5.358307556390321e-05, 'learning_rate/base': 1.3395768890975802e-05, 'epoch': 0.42}) +[2026-06-30 00:24:31,849][starVLA.training.train_starvla][INFO] - Step 980, Loss: {'train/loss': 0.009236310608685017, 'train/grad_norm_pre_clip': 2.0142369270324707, 'timing/data': 0.00021795090287923813, 'timing/model': 1.2325959699228406, 'learning_rate/action_model': 5.350771643575016e-05, 'learning_rate/action_query_proj': 5.350771643575016e-05, 'learning_rate/base': 1.337692910893754e-05, 'epoch': 0.42}) +[2026-06-30 00:25:04,553][starVLA.training.train_starvla][INFO] - Step 981, Loss: {'train/loss': 0.04653094336390495, 'train/grad_norm_pre_clip': 1.1887552738189697, 'timing/data': 0.00030059716664254665, 'timing/model': 1.2319435649551451, 'learning_rate/action_model': 5.3432353587456683e-05, 'learning_rate/action_query_proj': 5.3432353587456683e-05, 'learning_rate/base': 1.3358088396864171e-05, 'epoch': 0.42}) +[2026-06-30 00:25:37,229][starVLA.training.train_starvla][INFO] - Step 982, Loss: {'train/loss': 7.423066563205793e-05, 'train/grad_norm_pre_clip': 3.0816829204559326, 'timing/data': 0.0003491679672151804, 'timing/model': 1.2313393170479685, 'learning_rate/action_model': 5.335698720497313e-05, 'learning_rate/action_query_proj': 5.335698720497313e-05, 'learning_rate/base': 1.3339246801243283e-05, 'epoch': 0.42}) +[2026-06-30 00:26:09,926][starVLA.training.train_starvla][INFO] - Step 983, Loss: {'train/loss': 0.05333007127046585, 'train/grad_norm_pre_clip': 3.5045206546783447, 'timing/data': 0.0002601509913802147, 'timing/model': 1.2304907359648496, 'learning_rate/action_model': 5.3281617474258526e-05, 'learning_rate/action_query_proj': 5.3281617474258526e-05, 'learning_rate/base': 1.3320404368564631e-05, 'epoch': 0.42}) +[2026-06-30 00:26:42,620][starVLA.training.train_starvla][INFO] - Step 984, Loss: {'train/loss': 0.005955991335213184, 'train/grad_norm_pre_clip': 4.765608787536621, 'timing/data': 0.00020427489653229713, 'timing/model': 1.2275810760911554, 'learning_rate/action_model': 5.320624458128021e-05, 'learning_rate/action_query_proj': 5.320624458128021e-05, 'learning_rate/base': 1.3301561145320052e-05, 'epoch': 0.42}) +[2026-06-30 00:27:15,283][starVLA.training.train_starvla][INFO] - Step 985, Loss: {'train/loss': 0.3131673038005829, 'train/grad_norm_pre_clip': 3.078538656234741, 'timing/data': 0.00023850006982684135, 'timing/model': 1.2302570578176528, 'learning_rate/action_model': 5.3130868712013296e-05, 'learning_rate/action_query_proj': 5.3130868712013296e-05, 'learning_rate/base': 1.3282717178003324e-05, 'epoch': 0.42}) +[2026-06-30 00:27:47,992][starVLA.training.train_starvla][INFO] - Step 986, Loss: {'train/loss': 1.677842556091491e-05, 'train/grad_norm_pre_clip': 1.8225016593933105, 'timing/data': 0.0002822119276970625, 'timing/model': 1.2395093629602343, 'learning_rate/action_model': 5.305549005244025e-05, 'learning_rate/action_query_proj': 5.305549005244025e-05, 'learning_rate/base': 1.3263872513110063e-05, 'epoch': 0.42}) +[2026-06-30 00:28:20,698][starVLA.training.train_starvla][INFO] - Step 987, Loss: {'train/loss': 0.19231462478637695, 'train/grad_norm_pre_clip': 3.0409374237060547, 'timing/data': 0.00033300905488431454, 'timing/model': 1.2301674408372492, 'learning_rate/action_model': 5.298010878855043e-05, 'learning_rate/action_query_proj': 5.298010878855043e-05, 'learning_rate/base': 1.3245027197137608e-05, 'epoch': 0.42}) +[2026-06-30 00:28:53,407][starVLA.training.train_starvla][INFO] - Step 988, Loss: {'train/loss': 0.0026798956096172333, 'train/grad_norm_pre_clip': 1.5662147998809814, 'timing/data': 0.00035754404962062836, 'timing/model': 1.231674832990393, 'learning_rate/action_model': 5.29047251063396e-05, 'learning_rate/action_query_proj': 5.29047251063396e-05, 'learning_rate/base': 1.32261812765849e-05, 'epoch': 0.43}) +[2026-06-30 00:29:26,129][starVLA.training.train_starvla][INFO] - Step 989, Loss: {'train/loss': 0.007294654380530119, 'train/grad_norm_pre_clip': 3.607553482055664, 'timing/data': 0.00019519799388945103, 'timing/model': 1.2276616001036018, 'learning_rate/action_model': 5.28293391918095e-05, 'learning_rate/action_query_proj': 5.28293391918095e-05, 'learning_rate/base': 1.3207334797952375e-05, 'epoch': 0.43}) +[2026-06-30 00:29:58,980][starVLA.training.train_starvla][INFO] - Step 990, Loss: {'train/loss': 0.27547919750213623, 'train/grad_norm_pre_clip': 3.385547161102295, 'timing/data': 0.00025157397612929344, 'timing/model': 1.2341225631535053, 'learning_rate/action_model': 5.27539512309674e-05, 'learning_rate/action_query_proj': 5.27539512309674e-05, 'learning_rate/base': 1.318848780774185e-05, 'epoch': 0.43}) +[2026-06-30 00:30:31,705][starVLA.training.train_starvla][INFO] - Step 991, Loss: {'train/loss': 0.0013794959522783756, 'train/grad_norm_pre_clip': 2.1231689453125, 'timing/data': 0.00019751209765672684, 'timing/model': 1.2335910140536726, 'learning_rate/action_model': 5.2678561409825576e-05, 'learning_rate/action_query_proj': 5.2678561409825576e-05, 'learning_rate/base': 1.3169640352456394e-05, 'epoch': 0.43}) +[2026-06-30 00:31:04,394][starVLA.training.train_starvla][INFO] - Step 992, Loss: {'train/loss': 7.86388773121871e-05, 'train/grad_norm_pre_clip': 2.1745426654815674, 'timing/data': 0.00019407598301768303, 'timing/model': 1.2323835289571434, 'learning_rate/action_model': 5.260316991440093e-05, 'learning_rate/action_query_proj': 5.260316991440093e-05, 'learning_rate/base': 1.3150792478600232e-05, 'epoch': 0.43}) +[2026-06-30 00:31:37,066][starVLA.training.train_starvla][INFO] - Step 993, Loss: {'train/loss': 0.1573030799627304, 'train/grad_norm_pre_clip': 1.3563249111175537, 'timing/data': 0.0002770929131656885, 'timing/model': 1.2287746220827103, 'learning_rate/action_model': 5.252777693071448e-05, 'learning_rate/action_query_proj': 5.252777693071448e-05, 'learning_rate/base': 1.313194423267862e-05, 'epoch': 0.43}) +[2026-06-30 00:32:09,738][starVLA.training.train_starvla][INFO] - Step 994, Loss: {'train/loss': 7.640673720743507e-05, 'train/grad_norm_pre_clip': 1.977596402168274, 'timing/data': 0.0002987929619848728, 'timing/model': 1.2304730659816414, 'learning_rate/action_model': 5.245238264479092e-05, 'learning_rate/action_query_proj': 5.245238264479092e-05, 'learning_rate/base': 1.311309566119773e-05, 'epoch': 0.43}) +[2026-06-30 00:32:42,341][starVLA.training.train_starvla][INFO] - Step 995, Loss: {'train/loss': 3.707339055836201e-05, 'train/grad_norm_pre_clip': 1.8474688529968262, 'timing/data': 0.00024617393501102924, 'timing/model': 1.2293127349112183, 'learning_rate/action_model': 5.237698724265815e-05, 'learning_rate/action_query_proj': 5.237698724265815e-05, 'learning_rate/base': 1.3094246810664537e-05, 'epoch': 0.43}) +[2026-06-30 00:33:15,046][starVLA.training.train_starvla][INFO] - Step 996, Loss: {'train/loss': 0.47634175419807434, 'train/grad_norm_pre_clip': 1.8192238807678223, 'timing/data': 0.00021267100237309933, 'timing/model': 1.2324317859020084, 'learning_rate/action_model': 5.230159091034683e-05, 'learning_rate/action_query_proj': 5.230159091034683e-05, 'learning_rate/base': 1.3075397727586708e-05, 'epoch': 0.43}) +[2026-06-30 00:33:47,736][starVLA.training.train_starvla][INFO] - Step 997, Loss: {'train/loss': 0.0005198977305553854, 'train/grad_norm_pre_clip': 1.8039052486419678, 'timing/data': 0.0002900380641222, 'timing/model': 1.2332024050410837, 'learning_rate/action_model': 5.222619383388992e-05, 'learning_rate/action_query_proj': 5.222619383388992e-05, 'learning_rate/base': 1.305654845847248e-05, 'epoch': 0.43}) +[2026-06-30 00:34:20,383][starVLA.training.train_starvla][INFO] - Step 998, Loss: {'train/loss': 0.11447720974683762, 'train/grad_norm_pre_clip': 2.284404993057251, 'timing/data': 0.00030029681511223316, 'timing/model': 1.2302545888815075, 'learning_rate/action_model': 5.215079619932222e-05, 'learning_rate/action_query_proj': 5.215079619932222e-05, 'learning_rate/base': 1.3037699049830555e-05, 'epoch': 0.43}) +[2026-06-30 00:34:52,995][starVLA.training.train_starvla][INFO] - Step 999, Loss: {'train/loss': 0.022867081686854362, 'train/grad_norm_pre_clip': 2.1451640129089355, 'timing/data': 0.00021138903684914112, 'timing/model': 1.232985015027225, 'learning_rate/action_model': 5.207539819267988e-05, 'learning_rate/action_query_proj': 5.207539819267988e-05, 'learning_rate/base': 1.301884954816997e-05, 'epoch': 0.43}) +[2026-06-30 00:35:25,716][starVLA.training.train_starvla][INFO] - Starting held-out action loss eval at step 1000 +[2026-06-30 00:36:03,064][starVLA.training.train_starvla][INFO] - Finished held-out action loss eval at step 1000 +[2026-06-30 00:36:03,064][starVLA.training.train_starvla][INFO] - Starting held-out action classification eval at step 1000 +[2026-06-30 00:36:03,065][starVLA.training.train_starvla][INFO] - Finished held-out action classification eval at step 1000 +[2026-06-30 01:24:39,541][starVLA.training.rl_games.checkpoint_sync][INFO] - HF checkpoint sync disabled; skipping eval result upload for step 1000 +[2026-06-30 01:24:39,542][starVLA.training.train_starvla][INFO] - Step 1000, Loss: {'train/loss': 0.5741031765937805, 'train/grad_norm_pre_clip': 2.9910411834716797, 'eval/action_loss/seconds': 37.34551316080615, 'eval/action_loss/samples': 200.0, 'eval/loss': 0.046862758696079254, 'eval/latency_1/loss': 0.046862758696079254, 'eval/flappy/loss': 0.046862758696079254, 'eval/flappy/latency_1/loss': 0.046862758696079254, 'rl_games_eval/mid_train/total_episodes': 5.0, 'rl_games_eval/mid_train/mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/mean_length': 3600.0, 'rl_games_eval/mid_train/std_reward': 0.0, 'rl_games_eval/mid_train/std_length': 0.0, 'rl_games_eval/mid_train/macro_mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/macro_mean_length': 3600.0, 'rl_games_eval/mid_train/task_count': 1.0, 'rl_games_eval/mid_train/flappy__latency_1/mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/flappy__latency_1/mean_length': 3600.0, 'rl_games_eval/mid_train/flappy__latency_1/std_reward': 0.0, 'rl_games_eval/mid_train/flappy__latency_1/std_length': 0.0, 'rl_games_eval/mid_train/flappy__latency_1/num_episodes': 5.0, 'checkpoint/current_eval_score': 444.60000000002566, 'checkpoint/best_score': -inf, 'checkpoint/best_step': 0.0, 'checkpoint/is_new_best': 0.0, 'timing/data': 0.00030210986733436584, 'timing/model': 1.2365049680229276, 'learning_rate/action_model': 5.2000000000000004e-05, 'learning_rate/action_query_proj': 5.2000000000000004e-05, 'learning_rate/base': 1.3000000000000001e-05, 'epoch': 0.43}) +[2026-06-30 01:25:12,220][starVLA.training.train_starvla][INFO] - Step 1001, Loss: {'train/loss': 0.03240352123975754, 'train/grad_norm_pre_clip': 4.651487827301025, 'timing/data': 0.0005082790739834309, 'timing/model': 1.2332195090129972, 'learning_rate/action_model': 5.192460180732013e-05, 'learning_rate/action_query_proj': 5.192460180732013e-05, 'learning_rate/base': 1.2981150451830032e-05, 'epoch': 0.43}) +[2026-06-30 01:25:44,890][starVLA.training.train_starvla][INFO] - Step 1002, Loss: {'train/loss': 0.00014541247219312936, 'train/grad_norm_pre_clip': 2.512446403503418, 'timing/data': 0.00025787693448364735, 'timing/model': 1.2322966668289155, 'learning_rate/action_model': 5.184920380067778e-05, 'learning_rate/action_query_proj': 5.184920380067778e-05, 'learning_rate/base': 1.2962300950169445e-05, 'epoch': 0.43}) +[2026-06-30 01:26:17,591][starVLA.training.train_starvla][INFO] - Step 1003, Loss: {'train/loss': 3.895007830578834e-05, 'train/grad_norm_pre_clip': 2.708796262741089, 'timing/data': 0.00020900415256619453, 'timing/model': 1.2357356040738523, 'learning_rate/action_model': 5.1773806166110085e-05, 'learning_rate/action_query_proj': 5.1773806166110085e-05, 'learning_rate/base': 1.2943451541527521e-05, 'epoch': 0.43}) +[2026-06-30 01:26:50,249][starVLA.training.train_starvla][INFO] - Step 1004, Loss: {'train/loss': 0.5032259225845337, 'train/grad_norm_pre_clip': 1.7900482416152954, 'timing/data': 0.0004310738295316696, 'timing/model': 1.2313048830255866, 'learning_rate/action_model': 5.1698409089653177e-05, 'learning_rate/action_query_proj': 5.1698409089653177e-05, 'learning_rate/base': 1.2924602272413294e-05, 'epoch': 0.43}) +[2026-06-30 01:27:22,946][starVLA.training.train_starvla][INFO] - Step 1005, Loss: {'train/loss': 0.4047411382198334, 'train/grad_norm_pre_clip': 3.802333354949951, 'timing/data': 0.00020811217837035656, 'timing/model': 1.2292019431479275, 'learning_rate/action_model': 5.162301275734187e-05, 'learning_rate/action_query_proj': 5.162301275734187e-05, 'learning_rate/base': 1.2905753189335468e-05, 'epoch': 0.43}) +[2026-06-30 01:27:55,624][starVLA.training.train_starvla][INFO] - Step 1006, Loss: {'train/loss': 7.163699774537235e-05, 'train/grad_norm_pre_clip': 5.366022109985352, 'timing/data': 0.0003511919640004635, 'timing/model': 1.2310672421008348, 'learning_rate/action_model': 5.154761735520909e-05, 'learning_rate/action_query_proj': 5.154761735520909e-05, 'learning_rate/base': 1.2886904338802272e-05, 'epoch': 0.43}) +[2026-06-30 01:28:28,283][starVLA.training.train_starvla][INFO] - Step 1007, Loss: {'train/loss': 0.000797593267634511, 'train/grad_norm_pre_clip': 3.2861781120300293, 'timing/data': 0.0002693380229175091, 'timing/model': 1.232852339046076, 'learning_rate/action_model': 5.147222306928554e-05, 'learning_rate/action_query_proj': 5.147222306928554e-05, 'learning_rate/base': 1.2868055767321385e-05, 'epoch': 0.43}) +[2026-06-30 01:29:00,941][starVLA.training.train_starvla][INFO] - Step 1008, Loss: {'train/loss': 0.001076593529433012, 'train/grad_norm_pre_clip': 2.736182689666748, 'timing/data': 0.00025354791432619095, 'timing/model': 1.2303905528970063, 'learning_rate/action_model': 5.139683008559907e-05, 'learning_rate/action_query_proj': 5.139683008559907e-05, 'learning_rate/base': 1.2849207521399767e-05, 'epoch': 0.43}) +[2026-06-30 01:29:33,601][starVLA.training.train_starvla][INFO] - Step 1009, Loss: {'train/loss': 0.007640909403562546, 'train/grad_norm_pre_clip': 1.251902461051941, 'timing/data': 0.000201330054551363, 'timing/model': 1.2293600579723716, 'learning_rate/action_model': 5.1321438590174445e-05, 'learning_rate/action_query_proj': 5.1321438590174445e-05, 'learning_rate/base': 1.2830359647543611e-05, 'epoch': 0.43}) +[2026-06-30 01:30:06,248][starVLA.training.train_starvla][INFO] - Step 1010, Loss: {'train/loss': 0.015040544793009758, 'train/grad_norm_pre_clip': 2.0757458209991455, 'timing/data': 0.00024841795675456524, 'timing/model': 1.228470227913931, 'learning_rate/action_model': 5.124604876903262e-05, 'learning_rate/action_query_proj': 5.124604876903262e-05, 'learning_rate/base': 1.2811512192258155e-05, 'epoch': 0.43}) +[2026-06-30 01:30:38,915][starVLA.training.train_starvla][INFO] - Step 1011, Loss: {'train/loss': 0.00048444975982420146, 'train/grad_norm_pre_clip': 2.6462323665618896, 'timing/data': 0.00020220200531184673, 'timing/model': 1.2290554021019489, 'learning_rate/action_model': 5.1170660808190515e-05, 'learning_rate/action_query_proj': 5.1170660808190515e-05, 'learning_rate/base': 1.2792665202047629e-05, 'epoch': 0.44}) +[2026-06-30 01:31:11,644][starVLA.training.train_starvla][INFO] - Step 1012, Loss: {'train/loss': 0.16128358244895935, 'train/grad_norm_pre_clip': 4.356062412261963, 'timing/data': 0.00021276087500154972, 'timing/model': 1.236991835059598, 'learning_rate/action_model': 5.109527489366042e-05, 'learning_rate/action_query_proj': 5.109527489366042e-05, 'learning_rate/base': 1.2773818723415104e-05, 'epoch': 0.44}) +[2026-06-30 01:31:44,288][starVLA.training.train_starvla][INFO] - Step 1013, Loss: {'train/loss': 0.03302063047885895, 'train/grad_norm_pre_clip': 3.8771183490753174, 'timing/data': 0.000334311043843627, 'timing/model': 1.2308878710027784, 'learning_rate/action_model': 5.101989121144959e-05, 'learning_rate/action_query_proj': 5.101989121144959e-05, 'learning_rate/base': 1.2754972802862397e-05, 'epoch': 0.44}) +[2026-06-30 01:32:16,990][starVLA.training.train_starvla][INFO] - Step 1014, Loss: {'train/loss': 0.0012167423265054822, 'train/grad_norm_pre_clip': 1.4008898735046387, 'timing/data': 0.00022787903435528278, 'timing/model': 1.2363547419663519, 'learning_rate/action_model': 5.094450994755976e-05, 'learning_rate/action_query_proj': 5.094450994755976e-05, 'learning_rate/base': 1.273612748688994e-05, 'epoch': 0.44}) +[2026-06-30 01:32:49,655][starVLA.training.train_starvla][INFO] - Step 1015, Loss: {'train/loss': 0.47861647605895996, 'train/grad_norm_pre_clip': 4.006461143493652, 'timing/data': 0.00021992484107613564, 'timing/model': 1.2279360629618168, 'learning_rate/action_model': 5.0869131287986725e-05, 'learning_rate/action_query_proj': 5.0869131287986725e-05, 'learning_rate/base': 1.2717282821996681e-05, 'epoch': 0.44}) +[2026-06-30 01:33:22,320][starVLA.training.train_starvla][INFO] - Step 1016, Loss: {'train/loss': 0.01779301092028618, 'train/grad_norm_pre_clip': 2.7145543098449707, 'timing/data': 0.000199577072635293, 'timing/model': 1.228882985189557, 'learning_rate/action_model': 5.07937554187198e-05, 'learning_rate/action_query_proj': 5.07937554187198e-05, 'learning_rate/base': 1.269843885467995e-05, 'epoch': 0.44}) +[2026-06-30 01:33:54,995][starVLA.training.train_starvla][INFO] - Step 1017, Loss: {'train/loss': 0.0008863388211466372, 'train/grad_norm_pre_clip': 4.3991312980651855, 'timing/data': 0.0002672639675438404, 'timing/model': 1.2298767229076475, 'learning_rate/action_model': 5.0718382525741495e-05, 'learning_rate/action_query_proj': 5.0718382525741495e-05, 'learning_rate/base': 1.2679595631435374e-05, 'epoch': 0.44}) +[2026-06-30 01:34:27,653][starVLA.training.train_starvla][INFO] - Step 1018, Loss: {'train/loss': 0.0013798674335703254, 'train/grad_norm_pre_clip': 1.7409327030181885, 'timing/data': 0.00030030705966055393, 'timing/model': 1.2304737309459597, 'learning_rate/action_model': 5.064301279502689e-05, 'learning_rate/action_query_proj': 5.064301279502689e-05, 'learning_rate/base': 1.2660753198756722e-05, 'epoch': 0.44}) +[2026-06-30 01:35:00,315][starVLA.training.train_starvla][INFO] - Step 1019, Loss: {'train/loss': 0.016437791287899017, 'train/grad_norm_pre_clip': 3.4509525299072266, 'timing/data': 0.0002078020479530096, 'timing/model': 1.2296475919429213, 'learning_rate/action_model': 5.0567646412543324e-05, 'learning_rate/action_query_proj': 5.0567646412543324e-05, 'learning_rate/base': 1.2641911603135831e-05, 'epoch': 0.44}) +[2026-06-30 01:35:32,974][starVLA.training.train_starvla][INFO] - Step 1020, Loss: {'train/loss': 0.0002869307645596564, 'train/grad_norm_pre_clip': 4.794682025909424, 'timing/data': 0.00020230095833539963, 'timing/model': 1.231569966999814, 'learning_rate/action_model': 5.0492283564249844e-05, 'learning_rate/action_query_proj': 5.0492283564249844e-05, 'learning_rate/base': 1.2623070891062461e-05, 'epoch': 0.44}) +[2026-06-30 01:36:05,604][starVLA.training.train_starvla][INFO] - Step 1021, Loss: {'train/loss': 0.10410642623901367, 'train/grad_norm_pre_clip': 1.5802637338638306, 'timing/data': 0.00019330414943397045, 'timing/model': 1.2284022569656372, 'learning_rate/action_model': 5.0416924436096804e-05, 'learning_rate/action_query_proj': 5.0416924436096804e-05, 'learning_rate/base': 1.2604231109024201e-05, 'epoch': 0.44}) +[2026-06-30 01:36:38,244][starVLA.training.train_starvla][INFO] - Step 1022, Loss: {'train/loss': 3.0933744710637257e-05, 'train/grad_norm_pre_clip': 3.5784716606140137, 'timing/data': 0.00024184607900679111, 'timing/model': 1.2294384359847754, 'learning_rate/action_model': 5.034156921402533e-05, 'learning_rate/action_query_proj': 5.034156921402533e-05, 'learning_rate/base': 1.2585392303506333e-05, 'epoch': 0.44}) +[2026-06-30 01:37:10,870][starVLA.training.train_starvla][INFO] - Step 1023, Loss: {'train/loss': 0.40741533041000366, 'train/grad_norm_pre_clip': 6.102288246154785, 'timing/data': 0.0003460231237113476, 'timing/model': 1.2257166090421379, 'learning_rate/action_model': 5.0266218083966985e-05, 'learning_rate/action_query_proj': 5.0266218083966985e-05, 'learning_rate/base': 1.2566554520991746e-05, 'epoch': 0.44}) +[2026-06-30 01:37:43,484][starVLA.training.train_starvla][INFO] - Step 1024, Loss: {'train/loss': 7.226335583254695e-05, 'train/grad_norm_pre_clip': 2.6380832195281982, 'timing/data': 0.0002584070898592472, 'timing/model': 1.2273607219103724, 'learning_rate/action_model': 5.019087123184314e-05, 'learning_rate/action_query_proj': 5.019087123184314e-05, 'learning_rate/base': 1.2547717807960784e-05, 'epoch': 0.44}) +[2026-06-30 01:38:16,202][starVLA.training.train_starvla][INFO] - Step 1025, Loss: {'train/loss': 0.013756167143583298, 'train/grad_norm_pre_clip': 1.8633185625076294, 'timing/data': 0.00029843300580978394, 'timing/model': 1.2300547941122204, 'learning_rate/action_model': 5.011552884356472e-05, 'learning_rate/action_query_proj': 5.011552884356472e-05, 'learning_rate/base': 1.252888221089118e-05, 'epoch': 0.44}) +[2026-06-30 01:38:48,862][starVLA.training.train_starvla][INFO] - Step 1026, Loss: {'train/loss': 0.0007669363403692842, 'train/grad_norm_pre_clip': 3.113893747329712, 'timing/data': 0.0002744179219007492, 'timing/model': 1.2314641817938536, 'learning_rate/action_model': 5.004019110503152e-05, 'learning_rate/action_query_proj': 5.004019110503152e-05, 'learning_rate/base': 1.251004777625788e-05, 'epoch': 0.44}) +[2026-06-30 01:39:21,548][starVLA.training.train_starvla][INFO] - Step 1027, Loss: {'train/loss': 0.00018375938816461712, 'train/grad_norm_pre_clip': 1.7863032817840576, 'timing/data': 0.0002029919996857643, 'timing/model': 1.2289413399994373, 'learning_rate/action_model': 4.996485820213197e-05, 'learning_rate/action_query_proj': 4.996485820213197e-05, 'learning_rate/base': 1.2491214550532993e-05, 'epoch': 0.44}) +[2026-06-30 01:39:54,235][starVLA.training.train_starvla][INFO] - Step 1028, Loss: {'train/loss': 0.5919318795204163, 'train/grad_norm_pre_clip': 3.5236289501190186, 'timing/data': 0.00029358407482504845, 'timing/model': 1.2374689439311624, 'learning_rate/action_model': 4.988953032074249e-05, 'learning_rate/action_query_proj': 4.988953032074249e-05, 'learning_rate/base': 1.2472382580185622e-05, 'epoch': 0.44}) +[2026-06-30 01:40:26,902][starVLA.training.train_starvla][INFO] - Step 1029, Loss: {'train/loss': 0.04690547287464142, 'train/grad_norm_pre_clip': 2.0564138889312744, 'timing/data': 0.0002026129513978958, 'timing/model': 1.2321597340051085, 'learning_rate/action_model': 4.981420764672715e-05, 'learning_rate/action_query_proj': 4.981420764672715e-05, 'learning_rate/base': 1.2453551911681788e-05, 'epoch': 0.44}) +[2026-06-30 01:40:59,584][starVLA.training.train_starvla][INFO] - Step 1030, Loss: {'train/loss': 6.582698551937938e-05, 'train/grad_norm_pre_clip': 1.2082850933074951, 'timing/data': 0.00024813786149024963, 'timing/model': 1.2294682909268886, 'learning_rate/action_model': 4.973889036593715e-05, 'learning_rate/action_query_proj': 4.973889036593715e-05, 'learning_rate/base': 1.2434722591484288e-05, 'epoch': 0.44}) +[2026-06-30 01:41:32,276][starVLA.training.train_starvla][INFO] - Step 1031, Loss: {'train/loss': 0.5774928331375122, 'train/grad_norm_pre_clip': 2.1809422969818115, 'timing/data': 0.0002120397984981537, 'timing/model': 1.2282097518909723, 'learning_rate/action_model': 4.9663578664210416e-05, 'learning_rate/action_query_proj': 4.9663578664210416e-05, 'learning_rate/base': 1.2415894666052604e-05, 'epoch': 0.44}) +[2026-06-30 01:42:04,947][starVLA.training.train_starvla][INFO] - Step 1032, Loss: {'train/loss': 0.05304429680109024, 'train/grad_norm_pre_clip': 3.419776678085327, 'timing/data': 0.0003174189478158951, 'timing/model': 1.2292010050732642, 'learning_rate/action_model': 4.958827272737107e-05, 'learning_rate/action_query_proj': 4.958827272737107e-05, 'learning_rate/base': 1.2397068181842767e-05, 'epoch': 0.44}) +[2026-06-30 01:42:37,610][starVLA.training.train_starvla][INFO] - Step 1033, Loss: {'train/loss': 0.02890951558947563, 'train/grad_norm_pre_clip': 3.812932252883911, 'timing/data': 0.0002175401896238327, 'timing/model': 1.229207729920745, 'learning_rate/action_model': 4.951297274122902e-05, 'learning_rate/action_query_proj': 4.951297274122902e-05, 'learning_rate/base': 1.2378243185307255e-05, 'epoch': 0.44}) +[2026-06-30 01:43:10,263][starVLA.training.train_starvla][INFO] - Step 1034, Loss: {'train/loss': 0.17925527691841125, 'train/grad_norm_pre_clip': 3.089076042175293, 'timing/data': 0.0002112679649144411, 'timing/model': 1.2299551509786397, 'learning_rate/action_model': 4.94376788915795e-05, 'learning_rate/action_query_proj': 4.94376788915795e-05, 'learning_rate/base': 1.2359419722894874e-05, 'epoch': 0.44}) +[2026-06-30 01:43:42,955][starVLA.training.train_starvla][INFO] - Step 1035, Loss: {'train/loss': 0.04168320819735527, 'train/grad_norm_pre_clip': 2.352818012237549, 'timing/data': 0.00026061199605464935, 'timing/model': 1.2283084949012846, 'learning_rate/action_model': 4.9362391364202614e-05, 'learning_rate/action_query_proj': 4.9362391364202614e-05, 'learning_rate/base': 1.2340597841050654e-05, 'epoch': 0.45}) +[2026-06-30 01:44:15,620][starVLA.training.train_starvla][INFO] - Step 1036, Loss: {'train/loss': 0.3443385660648346, 'train/grad_norm_pre_clip': 2.7447752952575684, 'timing/data': 0.00019335513934493065, 'timing/model': 1.2335839471779764, 'learning_rate/action_model': 4.928711034486282e-05, 'learning_rate/action_query_proj': 4.928711034486282e-05, 'learning_rate/base': 1.2321777586215706e-05, 'epoch': 0.45}) +[2026-06-30 01:44:48,254][starVLA.training.train_starvla][INFO] - Step 1037, Loss: {'train/loss': 0.14762939512729645, 'train/grad_norm_pre_clip': 1.8948720693588257, 'timing/data': 0.0002511839848011732, 'timing/model': 1.2276501031592488, 'learning_rate/action_model': 4.921183601930859e-05, 'learning_rate/action_query_proj': 4.921183601930859e-05, 'learning_rate/base': 1.2302959004827148e-05, 'epoch': 0.45}) +[2026-06-30 01:45:20,952][starVLA.training.train_starvla][INFO] - Step 1038, Loss: {'train/loss': 0.03600335493683815, 'train/grad_norm_pre_clip': 1.638609528541565, 'timing/data': 0.00020111002959311008, 'timing/model': 1.2335881041362882, 'learning_rate/action_model': 4.913656857327181e-05, 'learning_rate/action_query_proj': 4.913656857327181e-05, 'learning_rate/base': 1.2284142143317952e-05, 'epoch': 0.45}) +[2026-06-30 01:45:53,809][starVLA.training.train_starvla][INFO] - Step 1039, Loss: {'train/loss': 0.0074751004576683044, 'train/grad_norm_pre_clip': 2.5749762058258057, 'timing/data': 0.0002131420187652111, 'timing/model': 1.23251061886549, 'learning_rate/action_model': 4.906130819246743e-05, 'learning_rate/action_query_proj': 4.906130819246743e-05, 'learning_rate/base': 1.2265327048116858e-05, 'epoch': 0.45}) +[2026-06-30 01:46:26,527][starVLA.training.train_starvla][INFO] - Step 1040, Loss: {'train/loss': 0.0003673369064927101, 'train/grad_norm_pre_clip': 2.6529853343963623, 'timing/data': 0.0002108379267156124, 'timing/model': 1.238020280841738, 'learning_rate/action_model': 4.8986055062592955e-05, 'learning_rate/action_query_proj': 4.8986055062592955e-05, 'learning_rate/base': 1.2246513765648239e-05, 'epoch': 0.45}) +[2026-06-30 01:46:59,260][starVLA.training.train_starvla][INFO] - Step 1041, Loss: {'train/loss': 0.011775564402341843, 'train/grad_norm_pre_clip': 2.015432119369507, 'timing/data': 0.00020074797794222832, 'timing/model': 1.2323821508325636, 'learning_rate/action_model': 4.891080936932802e-05, 'learning_rate/action_query_proj': 4.891080936932802e-05, 'learning_rate/base': 1.2227702342332005e-05, 'epoch': 0.45}) +[2026-06-30 01:47:31,978][starVLA.training.train_starvla][INFO] - Step 1042, Loss: {'train/loss': 0.18732447922229767, 'train/grad_norm_pre_clip': 2.703782081604004, 'timing/data': 0.0001949069555848837, 'timing/model': 1.2255276469513774, 'learning_rate/action_model': 4.883557129833386e-05, 'learning_rate/action_query_proj': 4.883557129833386e-05, 'learning_rate/base': 1.2208892824583465e-05, 'epoch': 0.45}) +[2026-06-30 01:48:04,662][starVLA.training.train_starvla][INFO] - Step 1043, Loss: {'train/loss': 0.02788008749485016, 'train/grad_norm_pre_clip': 1.6867358684539795, 'timing/data': 0.00019698217511177063, 'timing/model': 1.2309175820555538, 'learning_rate/action_model': 4.876034103525298e-05, 'learning_rate/action_query_proj': 4.876034103525298e-05, 'learning_rate/base': 1.2190085258813246e-05, 'epoch': 0.45}) +[2026-06-30 01:48:37,309][starVLA.training.train_starvla][INFO] - Step 1044, Loss: {'train/loss': 0.003122786059975624, 'train/grad_norm_pre_clip': 2.7344908714294434, 'timing/data': 0.00020260200835764408, 'timing/model': 1.2320514081511647, 'learning_rate/action_model': 4.868511876570852e-05, 'learning_rate/action_query_proj': 4.868511876570852e-05, 'learning_rate/base': 1.217127969142713e-05, 'epoch': 0.45}) +[2026-06-30 01:49:10,004][starVLA.training.train_starvla][INFO] - Step 1045, Loss: {'train/loss': 0.049251772463321686, 'train/grad_norm_pre_clip': 1.5950288772583008, 'timing/data': 0.0003550599794834852, 'timing/model': 1.2285104228649288, 'learning_rate/action_model': 4.8609904675304005e-05, 'learning_rate/action_query_proj': 4.8609904675304005e-05, 'learning_rate/base': 1.2152476168826001e-05, 'epoch': 0.45}) +[2026-06-30 01:49:42,695][starVLA.training.train_starvla][INFO] - Step 1046, Loss: {'train/loss': 0.0007614125497639179, 'train/grad_norm_pre_clip': 2.571537971496582, 'timing/data': 0.0001889062114059925, 'timing/model': 1.23278451198712, 'learning_rate/action_model': 4.853469894962269e-05, 'learning_rate/action_query_proj': 4.853469894962269e-05, 'learning_rate/base': 1.2133674737405672e-05, 'epoch': 0.45}) +[2026-06-30 01:50:15,360][starVLA.training.train_starvla][INFO] - Step 1047, Loss: {'train/loss': 0.05503755807876587, 'train/grad_norm_pre_clip': 1.0223835706710815, 'timing/data': 0.0002006480935961008, 'timing/model': 1.2331226540263742, 'learning_rate/action_model': 4.845950177422726e-05, 'learning_rate/action_query_proj': 4.845950177422726e-05, 'learning_rate/base': 1.2114875443556816e-05, 'epoch': 0.45}) +[2026-06-30 01:50:48,042][starVLA.training.train_starvla][INFO] - Step 1048, Loss: {'train/loss': 2.9801865821355022e-05, 'train/grad_norm_pre_clip': 1.8265234231948853, 'timing/data': 0.0002027831505984068, 'timing/model': 1.23059550113976, 'learning_rate/action_model': 4.838431333465924e-05, 'learning_rate/action_query_proj': 4.838431333465924e-05, 'learning_rate/base': 1.209607833366481e-05, 'epoch': 0.45}) +[2026-06-30 01:51:20,709][starVLA.training.train_starvla][INFO] - Step 1049, Loss: {'train/loss': 6.123959610704333e-05, 'train/grad_norm_pre_clip': 1.944668173789978, 'timing/data': 0.00025688507594168186, 'timing/model': 1.2295208980794996, 'learning_rate/action_model': 4.830913381643864e-05, 'learning_rate/action_query_proj': 4.830913381643864e-05, 'learning_rate/base': 1.207728345410966e-05, 'epoch': 0.45}) +[2026-06-30 01:51:53,393][starVLA.training.train_starvla][INFO] - Step 1050, Loss: {'train/loss': 0.010589329525828362, 'train/grad_norm_pre_clip': 3.290149450302124, 'timing/data': 0.00020293309353291988, 'timing/model': 1.2333880090154707, 'learning_rate/action_model': 4.8233963405063446e-05, 'learning_rate/action_query_proj': 4.8233963405063446e-05, 'learning_rate/base': 1.2058490851265861e-05, 'epoch': 0.45}) +[2026-06-30 01:52:26,110][starVLA.training.train_starvla][INFO] - Step 1051, Loss: {'train/loss': 0.009158330969512463, 'train/grad_norm_pre_clip': 2.1592791080474854, 'timing/data': 0.00019514909945428371, 'timing/model': 1.2403093101456761, 'learning_rate/action_model': 4.815880228600919e-05, 'learning_rate/action_query_proj': 4.815880228600919e-05, 'learning_rate/base': 1.2039700571502297e-05, 'epoch': 0.45}) +[2026-06-30 01:52:58,808][starVLA.training.train_starvla][INFO] - Step 1052, Loss: {'train/loss': 0.20630080997943878, 'train/grad_norm_pre_clip': 3.5387775897979736, 'timing/data': 0.0001955488696694374, 'timing/model': 1.2335881530307233, 'learning_rate/action_model': 4.808365064472844e-05, 'learning_rate/action_query_proj': 4.808365064472844e-05, 'learning_rate/base': 1.202091266118211e-05, 'epoch': 0.45}) +[2026-06-30 01:53:31,547][starVLA.training.train_starvla][INFO] - Step 1053, Loss: {'train/loss': 0.016842074692249298, 'train/grad_norm_pre_clip': 2.3798108100891113, 'timing/data': 0.00019231298938393593, 'timing/model': 1.2307547309901565, 'learning_rate/action_model': 4.800850866665042e-05, 'learning_rate/action_query_proj': 4.800850866665042e-05, 'learning_rate/base': 1.2002127166662605e-05, 'epoch': 0.45}) +[2026-06-30 01:54:04,242][starVLA.training.train_starvla][INFO] - Step 1054, Loss: {'train/loss': 0.4650431275367737, 'train/grad_norm_pre_clip': 6.073459625244141, 'timing/data': 0.00025465013459324837, 'timing/model': 1.2317516759503633, 'learning_rate/action_model': 4.7933376537180474e-05, 'learning_rate/action_query_proj': 4.7933376537180474e-05, 'learning_rate/base': 1.1983344134295118e-05, 'epoch': 0.45}) +[2026-06-30 01:54:36,919][starVLA.training.train_starvla][INFO] - Step 1055, Loss: {'train/loss': 0.7681216597557068, 'train/grad_norm_pre_clip': 2.202908992767334, 'timing/data': 0.0003546189982444048, 'timing/model': 1.2317194670904428, 'learning_rate/action_model': 4.7858254441699686e-05, 'learning_rate/action_query_proj': 4.7858254441699686e-05, 'learning_rate/base': 1.1964563610424921e-05, 'epoch': 0.45}) +[2026-06-30 01:55:09,568][starVLA.training.train_starvla][INFO] - Step 1056, Loss: {'train/loss': 0.013980300165712833, 'train/grad_norm_pre_clip': 2.25437331199646, 'timing/data': 0.00020460598170757294, 'timing/model': 1.2304450559895486, 'learning_rate/action_model': 4.778314256556433e-05, 'learning_rate/action_query_proj': 4.778314256556433e-05, 'learning_rate/base': 1.1945785641391082e-05, 'epoch': 0.45}) +[2026-06-30 01:55:42,210][starVLA.training.train_starvla][INFO] - Step 1057, Loss: {'train/loss': 5.2629107813118026e-05, 'train/grad_norm_pre_clip': 2.380570650100708, 'timing/data': 0.0002042150590568781, 'timing/model': 1.2298332140780985, 'learning_rate/action_model': 4.770804109410551e-05, 'learning_rate/action_query_proj': 4.770804109410551e-05, 'learning_rate/base': 1.1927010273526378e-05, 'epoch': 0.45}) +[2026-06-30 01:56:14,810][starVLA.training.train_starvla][INFO] - Step 1058, Loss: {'train/loss': 0.24793009459972382, 'train/grad_norm_pre_clip': 3.400850296020508, 'timing/data': 0.00020755082368850708, 'timing/model': 1.231895106844604, 'learning_rate/action_model': 4.763295021262864e-05, 'learning_rate/action_query_proj': 4.763295021262864e-05, 'learning_rate/base': 1.190823755315716e-05, 'epoch': 0.46}) +[2026-06-30 01:56:47,475][starVLA.training.train_starvla][INFO] - Step 1059, Loss: {'train/loss': 0.0007101725204847753, 'train/grad_norm_pre_clip': 2.974074602127075, 'timing/data': 0.00019685202278196812, 'timing/model': 1.2318578381091356, 'learning_rate/action_model': 4.755787010641302e-05, 'learning_rate/action_query_proj': 4.755787010641302e-05, 'learning_rate/base': 1.1889467526603255e-05, 'epoch': 0.46}) +[2026-06-30 01:57:20,167][starVLA.training.train_starvla][INFO] - Step 1060, Loss: {'train/loss': 0.0030484891030937433, 'train/grad_norm_pre_clip': 5.770419120788574, 'timing/data': 0.00040150806307792664, 'timing/model': 1.2308913171291351, 'learning_rate/action_model': 4.748280096071131e-05, 'learning_rate/action_query_proj': 4.748280096071131e-05, 'learning_rate/base': 1.1870700240177827e-05, 'epoch': 0.46}) +[2026-06-30 01:57:52,820][starVLA.training.train_starvla][INFO] - Step 1061, Loss: {'train/loss': 1.4626730680465698, 'train/grad_norm_pre_clip': 3.6185131072998047, 'timing/data': 0.00024988106451928616, 'timing/model': 1.2284175518434495, 'learning_rate/action_model': 4.740774296074921e-05, 'learning_rate/action_query_proj': 4.740774296074921e-05, 'learning_rate/base': 1.1851935740187303e-05, 'epoch': 0.46}) +[2026-06-30 01:58:25,385][starVLA.training.train_starvla][INFO] - Step 1062, Loss: {'train/loss': 0.0062201241962611675, 'train/grad_norm_pre_clip': 2.056607961654663, 'timing/data': 0.0002868310548365116, 'timing/model': 1.2380376940127462, 'learning_rate/action_model': 4.7332696291724826e-05, 'learning_rate/action_query_proj': 4.7332696291724826e-05, 'learning_rate/base': 1.1833174072931206e-05, 'epoch': 0.46}) +[2026-06-30 01:58:58,027][starVLA.training.train_starvla][INFO] - Step 1063, Loss: {'train/loss': 0.0001377935113850981, 'train/grad_norm_pre_clip': 1.2791755199432373, 'timing/data': 0.0003167979884892702, 'timing/model': 1.2315590330399573, 'learning_rate/action_model': 4.7257661138808405e-05, 'learning_rate/action_query_proj': 4.7257661138808405e-05, 'learning_rate/base': 1.1814415284702101e-05, 'epoch': 0.46}) +[2026-06-30 01:59:30,679][starVLA.training.train_starvla][INFO] - Step 1064, Loss: {'train/loss': 0.021531060338020325, 'train/grad_norm_pre_clip': 2.479461908340454, 'timing/data': 0.0001992359757423401, 'timing/model': 1.231256403028965, 'learning_rate/action_model': 4.718263768714168e-05, 'learning_rate/action_query_proj': 4.718263768714168e-05, 'learning_rate/base': 1.179565942178542e-05, 'epoch': 0.46}) +[2026-06-30 02:00:03,282][starVLA.training.train_starvla][INFO] - Step 1065, Loss: {'train/loss': 0.25653213262557983, 'train/grad_norm_pre_clip': 1.5343549251556396, 'timing/data': 0.0002519760746508837, 'timing/model': 1.2284855681937188, 'learning_rate/action_model': 4.7107626121837606e-05, 'learning_rate/action_query_proj': 4.7107626121837606e-05, 'learning_rate/base': 1.1776906530459401e-05, 'epoch': 0.46}) +[2026-06-30 02:00:35,962][starVLA.training.train_starvla][INFO] - Step 1066, Loss: {'train/loss': 0.0360407754778862, 'train/grad_norm_pre_clip': 1.6561298370361328, 'timing/data': 0.0002669640816748142, 'timing/model': 1.2298323400318623, 'learning_rate/action_model': 4.703262662797972e-05, 'learning_rate/action_query_proj': 4.703262662797972e-05, 'learning_rate/base': 1.175815665699493e-05, 'epoch': 0.46}) +[2026-06-30 02:01:08,665][starVLA.training.train_starvla][INFO] - Step 1067, Loss: {'train/loss': 0.05038932338356972, 'train/grad_norm_pre_clip': 1.586560845375061, 'timing/data': 0.0002991040237247944, 'timing/model': 1.2317792049143463, 'learning_rate/action_model': 4.6957639390621853e-05, 'learning_rate/action_query_proj': 4.6957639390621853e-05, 'learning_rate/base': 1.1739409847655463e-05, 'epoch': 0.46}) +[2026-06-30 02:01:41,341][starVLA.training.train_starvla][INFO] - Step 1068, Loss: {'train/loss': 0.2949088513851166, 'train/grad_norm_pre_clip': 3.4470114707946777, 'timing/data': 0.0002098260447382927, 'timing/model': 1.2297122061718255, 'learning_rate/action_model': 4.6882664594787524e-05, 'learning_rate/action_query_proj': 4.6882664594787524e-05, 'learning_rate/base': 1.1720666148696881e-05, 'epoch': 0.46}) +[2026-06-30 02:02:14,015][starVLA.training.train_starvla][INFO] - Step 1069, Loss: {'train/loss': 0.003939286805689335, 'train/grad_norm_pre_clip': 3.1953494548797607, 'timing/data': 0.00019875518046319485, 'timing/model': 1.233156014000997, 'learning_rate/action_model': 4.6807702425469624e-05, 'learning_rate/action_query_proj': 4.6807702425469624e-05, 'learning_rate/base': 1.1701925606367406e-05, 'epoch': 0.46}) +[2026-06-30 02:02:46,698][starVLA.training.train_starvla][INFO] - Step 1070, Loss: {'train/loss': 0.0004159801173955202, 'train/grad_norm_pre_clip': 3.9074113368988037, 'timing/data': 0.00020047789439558983, 'timing/model': 1.2265539499931037, 'learning_rate/action_model': 4.673275306762983e-05, 'learning_rate/action_query_proj': 4.673275306762983e-05, 'learning_rate/base': 1.1683188266907457e-05, 'epoch': 0.46}) +[2026-06-30 02:03:19,413][starVLA.training.train_starvla][INFO] - Step 1071, Loss: {'train/loss': 0.0012084313202649355, 'train/grad_norm_pre_clip': 2.3100016117095947, 'timing/data': 0.0002935640513896942, 'timing/model': 1.2371204760856926, 'learning_rate/action_model': 4.6657816706198246e-05, 'learning_rate/action_query_proj': 4.6657816706198246e-05, 'learning_rate/base': 1.1664454176549561e-05, 'epoch': 0.46}) +[2026-06-30 02:03:52,084][starVLA.training.train_starvla][INFO] - Step 1072, Loss: {'train/loss': 0.056276630610227585, 'train/grad_norm_pre_clip': 2.7886242866516113, 'timing/data': 0.00020724092610180378, 'timing/model': 1.2350534370634705, 'learning_rate/action_model': 4.658289352607288e-05, 'learning_rate/action_query_proj': 4.658289352607288e-05, 'learning_rate/base': 1.164572338151822e-05, 'epoch': 0.46}) +[2026-06-30 02:04:24,775][starVLA.training.train_starvla][INFO] - Step 1073, Loss: {'train/loss': 0.013277571648359299, 'train/grad_norm_pre_clip': 2.512439727783203, 'timing/data': 0.0002967000473290682, 'timing/model': 1.2314541379455477, 'learning_rate/action_model': 4.650798371211924e-05, 'learning_rate/action_query_proj': 4.650798371211924e-05, 'learning_rate/base': 1.162699592802981e-05, 'epoch': 0.46}) +[2026-06-30 02:04:57,419][starVLA.training.train_starvla][INFO] - Step 1074, Loss: {'train/loss': 4.359972081147134e-05, 'train/grad_norm_pre_clip': 2.984213352203369, 'timing/data': 0.00020997598767280579, 'timing/model': 1.228691547177732, 'learning_rate/action_model': 4.6433087449169855e-05, 'learning_rate/action_query_proj': 4.6433087449169855e-05, 'learning_rate/base': 1.1608271862292464e-05, 'epoch': 0.46}) +[2026-06-30 02:05:30,083][starVLA.training.train_starvla][INFO] - Step 1075, Loss: {'train/loss': 0.0002802839735522866, 'train/grad_norm_pre_clip': 1.7279777526855469, 'timing/data': 0.00043688411824405193, 'timing/model': 1.2184320210944861, 'learning_rate/action_model': 4.635820492202379e-05, 'learning_rate/action_query_proj': 4.635820492202379e-05, 'learning_rate/base': 1.1589551230505948e-05, 'epoch': 0.46}) +[2026-06-30 02:06:02,753][starVLA.training.train_starvla][INFO] - Step 1076, Loss: {'train/loss': 4.6936947910580784e-05, 'train/grad_norm_pre_clip': 1.824337124824524, 'timing/data': 0.00020764116197824478, 'timing/model': 1.2360417400486767, 'learning_rate/action_model': 4.628333631544625e-05, 'learning_rate/action_query_proj': 4.628333631544625e-05, 'learning_rate/base': 1.1570834078861562e-05, 'epoch': 0.46}) +[2026-06-30 02:06:35,489][starVLA.training.train_starvla][INFO] - Step 1077, Loss: {'train/loss': 0.009074878878891468, 'train/grad_norm_pre_clip': 4.262296676635742, 'timing/data': 0.00029348302632570267, 'timing/model': 1.2346554149407893, 'learning_rate/action_model': 4.620848181416808e-05, 'learning_rate/action_query_proj': 4.620848181416808e-05, 'learning_rate/base': 1.155212045354202e-05, 'epoch': 0.46}) +[2026-06-30 02:07:08,173][starVLA.training.train_starvla][INFO] - Step 1078, Loss: {'train/loss': 0.0004764854966197163, 'train/grad_norm_pre_clip': 1.9330790042877197, 'timing/data': 0.0001965709961950779, 'timing/model': 1.2297621020115912, 'learning_rate/action_model': 4.61336416028853e-05, 'learning_rate/action_query_proj': 4.61336416028853e-05, 'learning_rate/base': 1.1533410400721326e-05, 'epoch': 0.46}) +[2026-06-30 02:07:40,858][starVLA.training.train_starvla][INFO] - Step 1079, Loss: {'train/loss': 0.1601567417383194, 'train/grad_norm_pre_clip': 3.763701915740967, 'timing/data': 0.00020180991850793362, 'timing/model': 1.2305273287929595, 'learning_rate/action_model': 4.605881586625873e-05, 'learning_rate/action_query_proj': 4.605881586625873e-05, 'learning_rate/base': 1.1514703966564683e-05, 'epoch': 0.46}) +[2026-06-30 02:08:13,539][starVLA.training.train_starvla][INFO] - Step 1080, Loss: {'train/loss': 0.19570457935333252, 'train/grad_norm_pre_clip': 3.637559175491333, 'timing/data': 0.00028945598751306534, 'timing/model': 1.2335131531581283, 'learning_rate/action_model': 4.5984004788913385e-05, 'learning_rate/action_query_proj': 4.5984004788913385e-05, 'learning_rate/base': 1.1496001197228346e-05, 'epoch': 0.46}) +[2026-06-30 02:08:46,217][starVLA.training.train_starvla][INFO] - Step 1081, Loss: {'train/loss': 0.004692485556006432, 'train/grad_norm_pre_clip': 2.284689426422119, 'timing/data': 0.00020092912018299103, 'timing/model': 1.229843611130491, 'learning_rate/action_model': 4.590920855543823e-05, 'learning_rate/action_query_proj': 4.590920855543823e-05, 'learning_rate/base': 1.1477302138859557e-05, 'epoch': 0.47}) +[2026-06-30 02:09:18,840][starVLA.training.train_starvla][INFO] - Step 1082, Loss: {'train/loss': 0.04290285333991051, 'train/grad_norm_pre_clip': 1.4619557857513428, 'timing/data': 0.00029068905860185623, 'timing/model': 1.2308923969976604, 'learning_rate/action_model': 4.5834427350385455e-05, 'learning_rate/action_query_proj': 4.5834427350385455e-05, 'learning_rate/base': 1.1458606837596364e-05, 'epoch': 0.47}) +[2026-06-30 02:09:51,519][starVLA.training.train_starvla][INFO] - Step 1083, Loss: {'train/loss': 0.007892230525612831, 'train/grad_norm_pre_clip': 2.3603811264038086, 'timing/data': 0.00021015689708292484, 'timing/model': 1.2327310719992965, 'learning_rate/action_model': 4.575966135827033e-05, 'learning_rate/action_query_proj': 4.575966135827033e-05, 'learning_rate/base': 1.1439915339567582e-05, 'epoch': 0.47}) +[2026-06-30 02:10:24,170][starVLA.training.train_starvla][INFO] - Step 1084, Loss: {'train/loss': 0.01520696934312582, 'train/grad_norm_pre_clip': 1.7574864625930786, 'timing/data': 0.0001930040307343006, 'timing/model': 1.2293859261553735, 'learning_rate/action_model': 4.5684910763570435e-05, 'learning_rate/action_query_proj': 4.5684910763570435e-05, 'learning_rate/base': 1.1421227690892609e-05, 'epoch': 0.47}) +[2026-06-30 02:10:56,799][starVLA.training.train_starvla][INFO] - Step 1085, Loss: {'train/loss': 0.09517461806535721, 'train/grad_norm_pre_clip': 2.4172487258911133, 'timing/data': 0.00020411494188010693, 'timing/model': 1.2288885319139808, 'learning_rate/action_model': 4.5610175750725495e-05, 'learning_rate/action_query_proj': 4.5610175750725495e-05, 'learning_rate/base': 1.1402543937681374e-05, 'epoch': 0.47}) +[2026-06-30 02:11:29,464][starVLA.training.train_starvla][INFO] - Step 1086, Loss: {'train/loss': 0.023640673607587814, 'train/grad_norm_pre_clip': 2.8326497077941895, 'timing/data': 0.00029022712260484695, 'timing/model': 1.230195279000327, 'learning_rate/action_model': 4.5535456504136674e-05, 'learning_rate/action_query_proj': 4.5535456504136674e-05, 'learning_rate/base': 1.1383864126034169e-05, 'epoch': 0.47}) +[2026-06-30 02:12:02,151][starVLA.training.train_starvla][INFO] - Step 1087, Loss: {'train/loss': 0.004167918581515551, 'train/grad_norm_pre_clip': 3.6044976711273193, 'timing/data': 0.0003176589962095022, 'timing/model': 1.2322026051115245, 'learning_rate/action_model': 4.5460753208166324e-05, 'learning_rate/action_query_proj': 4.5460753208166324e-05, 'learning_rate/base': 1.1365188302041581e-05, 'epoch': 0.47}) +[2026-06-30 02:12:34,828][starVLA.training.train_starvla][INFO] - Step 1088, Loss: {'train/loss': 0.000848545110784471, 'train/grad_norm_pre_clip': 1.7678908109664917, 'timing/data': 0.0002063601277768612, 'timing/model': 1.2348905839025974, 'learning_rate/action_model': 4.538606604713738e-05, 'learning_rate/action_query_proj': 4.538606604713738e-05, 'learning_rate/base': 1.1346516511784345e-05, 'epoch': 0.47}) +[2026-06-30 02:13:07,510][starVLA.training.train_starvla][INFO] - Step 1089, Loss: {'train/loss': 0.02127363346517086, 'train/grad_norm_pre_clip': 2.0467824935913086, 'timing/data': 0.00020724092610180378, 'timing/model': 1.2309290198609233, 'learning_rate/action_model': 4.531139520533299e-05, 'learning_rate/action_query_proj': 4.531139520533299e-05, 'learning_rate/base': 1.1327848801333248e-05, 'epoch': 0.47}) +[2026-06-30 02:13:40,168][starVLA.training.train_starvla][INFO] - Step 1090, Loss: {'train/loss': 0.003389641409739852, 'train/grad_norm_pre_clip': 3.2679800987243652, 'timing/data': 0.00042117410339415073, 'timing/model': 1.231667549815029, 'learning_rate/action_model': 4.523674086699603e-05, 'learning_rate/action_query_proj': 4.523674086699603e-05, 'learning_rate/base': 1.1309185216749008e-05, 'epoch': 0.47}) +[2026-06-30 02:14:13,027][starVLA.training.train_starvla][INFO] - Step 1091, Loss: {'train/loss': 0.0005470601026900113, 'train/grad_norm_pre_clip': 2.017862558364868, 'timing/data': 0.0002650599926710129, 'timing/model': 1.2287569460459054, 'learning_rate/action_model': 4.516210321632868e-05, 'learning_rate/action_query_proj': 4.516210321632868e-05, 'learning_rate/base': 1.129052580408217e-05, 'epoch': 0.47}) +[2026-06-30 02:14:45,700][starVLA.training.train_starvla][INFO] - Step 1092, Loss: {'train/loss': 0.13404187560081482, 'train/grad_norm_pre_clip': 1.6675198078155518, 'timing/data': 0.00030021602287888527, 'timing/model': 1.2291211569681764, 'learning_rate/action_model': 4.5087482437491896e-05, 'learning_rate/action_query_proj': 4.5087482437491896e-05, 'learning_rate/base': 1.1271870609372974e-05, 'epoch': 0.47}) +[2026-06-30 02:15:18,378][starVLA.training.train_starvla][INFO] - Step 1093, Loss: {'train/loss': 0.14390310645103455, 'train/grad_norm_pre_clip': 3.022850513458252, 'timing/data': 0.0003241009544581175, 'timing/model': 1.2304501479957253, 'learning_rate/action_model': 4.5012878714605055e-05, 'learning_rate/action_query_proj': 4.5012878714605055e-05, 'learning_rate/base': 1.1253219678651264e-05, 'epoch': 0.47}) +[2026-06-30 02:15:51,074][starVLA.training.train_starvla][INFO] - Step 1094, Loss: {'train/loss': 0.00014441563689615577, 'train/grad_norm_pre_clip': 1.4828048944473267, 'timing/data': 0.00026671309024095535, 'timing/model': 1.2344093387946486, 'learning_rate/action_model': 4.4938292231745397e-05, 'learning_rate/action_query_proj': 4.4938292231745397e-05, 'learning_rate/base': 1.1234573057936349e-05, 'epoch': 0.47}) +[2026-06-30 02:16:23,752][starVLA.training.train_starvla][INFO] - Step 1095, Loss: {'train/loss': 0.0070013622753322124, 'train/grad_norm_pre_clip': 3.4622490406036377, 'timing/data': 0.0003354828804731369, 'timing/model': 1.2318871340248734, 'learning_rate/action_model': 4.4863723172947694e-05, 'learning_rate/action_query_proj': 4.4863723172947694e-05, 'learning_rate/base': 1.1215930793236923e-05, 'epoch': 0.47}) +[2026-06-30 02:16:56,399][starVLA.training.train_starvla][INFO] - Step 1096, Loss: {'train/loss': 0.0009826455498114228, 'train/grad_norm_pre_clip': 2.402987241744995, 'timing/data': 0.0002275488805025816, 'timing/model': 1.2287580731790513, 'learning_rate/action_model': 4.478917172220366e-05, 'learning_rate/action_query_proj': 4.478917172220366e-05, 'learning_rate/base': 1.1197292930550914e-05, 'epoch': 0.47}) +[2026-06-30 02:17:29,075][starVLA.training.train_starvla][INFO] - Step 1097, Loss: {'train/loss': 0.019139207899570465, 'train/grad_norm_pre_clip': 3.029259443283081, 'timing/data': 0.00019655097275972366, 'timing/model': 1.2262085541151464, 'learning_rate/action_model': 4.471463806346159e-05, 'learning_rate/action_query_proj': 4.471463806346159e-05, 'learning_rate/base': 1.1178659515865397e-05, 'epoch': 0.47}) +[2026-06-30 02:18:01,747][starVLA.training.train_starvla][INFO] - Step 1098, Loss: {'train/loss': 0.11347851157188416, 'train/grad_norm_pre_clip': 1.9468882083892822, 'timing/data': 0.00026609189808368683, 'timing/model': 1.233223136048764, 'learning_rate/action_model': 4.464012238062588e-05, 'learning_rate/action_query_proj': 4.464012238062588e-05, 'learning_rate/base': 1.116003059515647e-05, 'epoch': 0.47}) +[2026-06-30 02:18:34,412][starVLA.training.train_starvla][INFO] - Step 1099, Loss: {'train/loss': 1.4126165297057014e-05, 'train/grad_norm_pre_clip': 2.059312343597412, 'timing/data': 0.00025322684086859226, 'timing/model': 1.23103571892716, 'learning_rate/action_model': 4.456562485755659e-05, 'learning_rate/action_query_proj': 4.456562485755659e-05, 'learning_rate/base': 1.1141406214389148e-05, 'epoch': 0.47}) +[2026-06-30 02:19:07,093][starVLA.training.train_starvla][INFO] - Step 1100, Loss: {'train/loss': 0.0002471519401296973, 'train/grad_norm_pre_clip': 2.0855696201324463, 'timing/data': 0.0001971821766346693, 'timing/model': 1.2313182519283146, 'learning_rate/action_model': 4.449114567806891e-05, 'learning_rate/action_query_proj': 4.449114567806891e-05, 'learning_rate/base': 1.1122786419517228e-05, 'epoch': 0.47}) +[2026-06-30 02:19:39,768][starVLA.training.train_starvla][INFO] - Step 1101, Loss: {'train/loss': 0.0002835195045918226, 'train/grad_norm_pre_clip': 3.9737484455108643, 'timing/data': 0.0002789159771054983, 'timing/model': 1.2359733909834176, 'learning_rate/action_model': 4.441668502593286e-05, 'learning_rate/action_query_proj': 4.441668502593286e-05, 'learning_rate/base': 1.1104171256483215e-05, 'epoch': 0.47}) +[2026-06-30 02:20:12,484][starVLA.training.train_starvla][INFO] - Step 1102, Loss: {'train/loss': 0.18472906947135925, 'train/grad_norm_pre_clip': 3.1074306964874268, 'timing/data': 0.00026170420460402966, 'timing/model': 1.235008429037407, 'learning_rate/action_model': 4.434224308487264e-05, 'learning_rate/action_query_proj': 4.434224308487264e-05, 'learning_rate/base': 1.108556077121816e-05, 'epoch': 0.47}) +[2026-06-30 02:20:45,113][starVLA.training.train_starvla][INFO] - Step 1103, Loss: {'train/loss': 0.00024558758013881743, 'train/grad_norm_pre_clip': 1.6283332109451294, 'timing/data': 0.0003476659767329693, 'timing/model': 1.2311361508909613, 'learning_rate/action_model': 4.426782003856642e-05, 'learning_rate/action_query_proj': 4.426782003856642e-05, 'learning_rate/base': 1.1066955009641604e-05, 'epoch': 0.47}) +[2026-06-30 02:21:17,786][starVLA.training.train_starvla][INFO] - Step 1104, Loss: {'train/loss': 0.041925329715013504, 'train/grad_norm_pre_clip': 2.9324467182159424, 'timing/data': 0.00019800406880676746, 'timing/model': 1.2303475711960346, 'learning_rate/action_model': 4.419341607064559e-05, 'learning_rate/action_query_proj': 4.419341607064559e-05, 'learning_rate/base': 1.1048354017661398e-05, 'epoch': 0.48}) +[2026-06-30 02:21:50,452][starVLA.training.train_starvla][INFO] - Step 1105, Loss: {'train/loss': 0.0026660251896828413, 'train/grad_norm_pre_clip': 1.268518328666687, 'timing/data': 0.00027206307277083397, 'timing/model': 1.2283336778637022, 'learning_rate/action_model': 4.411903136469458e-05, 'learning_rate/action_query_proj': 4.411903136469458e-05, 'learning_rate/base': 1.1029757841173646e-05, 'epoch': 0.48}) +[2026-06-30 02:22:23,086][starVLA.training.train_starvla][INFO] - Step 1106, Loss: {'train/loss': 0.04261026531457901, 'train/grad_norm_pre_clip': 2.1194276809692383, 'timing/data': 0.00020188000053167343, 'timing/model': 1.2287127189338207, 'learning_rate/action_model': 4.404466610425025e-05, 'learning_rate/action_query_proj': 4.404466610425025e-05, 'learning_rate/base': 1.1011166526062562e-05, 'epoch': 0.48}) +[2026-06-30 02:22:55,725][starVLA.training.train_starvla][INFO] - Step 1107, Loss: {'train/loss': 0.031176209449768066, 'train/grad_norm_pre_clip': 2.511648416519165, 'timing/data': 0.000248027965426445, 'timing/model': 1.2327170679345727, 'learning_rate/action_model': 4.39703204728015e-05, 'learning_rate/action_query_proj': 4.39703204728015e-05, 'learning_rate/base': 1.0992580118200375e-05, 'epoch': 0.48}) +[2026-06-30 02:23:28,369][starVLA.training.train_starvla][INFO] - Step 1108, Loss: {'train/loss': 0.0007918900228105485, 'train/grad_norm_pre_clip': 2.1953999996185303, 'timing/data': 0.000346834072843194, 'timing/model': 1.2346418958622962, 'learning_rate/action_model': 4.3895994653788775e-05, 'learning_rate/action_query_proj': 4.3895994653788775e-05, 'learning_rate/base': 1.0973998663447194e-05, 'epoch': 0.48}) +[2026-06-30 02:24:01,014][starVLA.training.train_starvla][INFO] - Step 1109, Loss: {'train/loss': 8.841325325192884e-05, 'train/grad_norm_pre_clip': 2.0877909660339355, 'timing/data': 0.00027679186314344406, 'timing/model': 1.2316914040129632, 'learning_rate/action_model': 4.382168883060365e-05, 'learning_rate/action_query_proj': 4.382168883060365e-05, 'learning_rate/base': 1.0955422207650912e-05, 'epoch': 0.48}) +[2026-06-30 02:24:33,702][starVLA.training.train_starvla][INFO] - Step 1110, Loss: {'train/loss': 6.88093132339418e-05, 'train/grad_norm_pre_clip': 3.495438575744629, 'timing/data': 0.00033267796970903873, 'timing/model': 1.2328227229882032, 'learning_rate/action_model': 4.3747403186588336e-05, 'learning_rate/action_query_proj': 4.3747403186588336e-05, 'learning_rate/base': 1.0936850796647084e-05, 'epoch': 0.48}) +[2026-06-30 02:25:06,366][starVLA.training.train_starvla][INFO] - Step 1111, Loss: {'train/loss': 0.0038897346239537, 'train/grad_norm_pre_clip': 1.6593884229660034, 'timing/data': 0.00020006787963211536, 'timing/model': 1.2286613369360566, 'learning_rate/action_model': 4.367313790503531e-05, 'learning_rate/action_query_proj': 4.367313790503531e-05, 'learning_rate/base': 1.0918284476258828e-05, 'epoch': 0.48}) +[2026-06-30 02:25:39,039][starVLA.training.train_starvla][INFO] - Step 1112, Loss: {'train/loss': 0.2737298607826233, 'train/grad_norm_pre_clip': 3.6774847507476807, 'timing/data': 0.00019717193208634853, 'timing/model': 1.2294612610712647, 'learning_rate/action_model': 4.359889316918674e-05, 'learning_rate/action_query_proj': 4.359889316918674e-05, 'learning_rate/base': 1.0899723292296685e-05, 'epoch': 0.48}) +[2026-06-30 02:26:11,716][starVLA.training.train_starvla][INFO] - Step 1113, Loss: {'train/loss': 0.4022179841995239, 'train/grad_norm_pre_clip': 3.0943329334259033, 'timing/data': 0.0001979528460651636, 'timing/model': 1.2365622310899198, 'learning_rate/action_model': 4.352466916223416e-05, 'learning_rate/action_query_proj': 4.352466916223416e-05, 'learning_rate/base': 1.088116729055854e-05, 'epoch': 0.48}) +[2026-06-30 02:26:44,400][starVLA.training.train_starvla][INFO] - Step 1114, Loss: {'train/loss': 0.02373071387410164, 'train/grad_norm_pre_clip': 1.7520005702972412, 'timing/data': 0.00020509702153503895, 'timing/model': 1.2342244840692729, 'learning_rate/action_model': 4.345046606731791e-05, 'learning_rate/action_query_proj': 4.345046606731791e-05, 'learning_rate/base': 1.0862616516829478e-05, 'epoch': 0.48}) +[2026-06-30 02:27:17,010][starVLA.training.train_starvla][INFO] - Step 1115, Loss: {'train/loss': 9.804902219912037e-06, 'train/grad_norm_pre_clip': 2.2538535594940186, 'timing/data': 0.0002813509199768305, 'timing/model': 1.2295512109994888, 'learning_rate/action_model': 4.337628406752676e-05, 'learning_rate/action_query_proj': 4.337628406752676e-05, 'learning_rate/base': 1.084407101688169e-05, 'epoch': 0.48}) +[2026-06-30 02:27:49,701][starVLA.training.train_starvla][INFO] - Step 1116, Loss: {'train/loss': 0.026998678222298622, 'train/grad_norm_pre_clip': 5.589297771453857, 'timing/data': 0.00019668089225888252, 'timing/model': 1.234236122108996, 'learning_rate/action_model': 4.33021233458974e-05, 'learning_rate/action_query_proj': 4.33021233458974e-05, 'learning_rate/base': 1.082553083647435e-05, 'epoch': 0.48}) +[2026-06-30 02:28:22,391][starVLA.training.train_starvla][INFO] - Step 1117, Loss: {'train/loss': 0.0005190433585084975, 'train/grad_norm_pre_clip': 2.2103610038757324, 'timing/data': 0.0002692979760468006, 'timing/model': 1.2249178159981966, 'learning_rate/action_model': 4.3227984085414074e-05, 'learning_rate/action_query_proj': 4.3227984085414074e-05, 'learning_rate/base': 1.0806996021353518e-05, 'epoch': 0.48}) +[2026-06-30 02:28:55,046][starVLA.training.train_starvla][INFO] - Step 1118, Loss: {'train/loss': 0.21304234862327576, 'train/grad_norm_pre_clip': 3.498049020767212, 'timing/data': 0.000297600869089365, 'timing/model': 1.2356795109808445, 'learning_rate/action_model': 4.315386646900799e-05, 'learning_rate/action_query_proj': 4.315386646900799e-05, 'learning_rate/base': 1.0788466617251998e-05, 'epoch': 0.48}) +[2026-06-30 02:29:27,692][starVLA.training.train_starvla][INFO] - Step 1119, Loss: {'train/loss': 0.00048176528071053326, 'train/grad_norm_pre_clip': 2.139350414276123, 'timing/data': 0.00020938506349921227, 'timing/model': 1.234516360098496, 'learning_rate/action_model': 4.3079770679557056e-05, 'learning_rate/action_query_proj': 4.3079770679557056e-05, 'learning_rate/base': 1.0769942669889264e-05, 'epoch': 0.48}) +[2026-06-30 02:30:00,287][starVLA.training.train_starvla][INFO] - Step 1120, Loss: {'train/loss': 0.1563558131456375, 'train/grad_norm_pre_clip': 1.143489122390747, 'timing/data': 0.00022822106257081032, 'timing/model': 1.2289259659592062, 'learning_rate/action_model': 4.3005696899885204e-05, 'learning_rate/action_query_proj': 4.3005696899885204e-05, 'learning_rate/base': 1.0751424224971301e-05, 'epoch': 0.48}) +[2026-06-30 02:30:32,960][starVLA.training.train_starvla][INFO] - Step 1121, Loss: {'train/loss': 0.026431774720549583, 'train/grad_norm_pre_clip': 2.5386953353881836, 'timing/data': 0.00024184607900679111, 'timing/model': 1.2339726779609919, 'learning_rate/action_model': 4.293164531276219e-05, 'learning_rate/action_query_proj': 4.293164531276219e-05, 'learning_rate/base': 1.0732911328190547e-05, 'epoch': 0.48}) +[2026-06-30 02:31:05,633][starVLA.training.train_starvla][INFO] - Step 1122, Loss: {'train/loss': 0.14302875101566315, 'train/grad_norm_pre_clip': 3.560957908630371, 'timing/data': 0.00022566597908735275, 'timing/model': 1.2308501300867647, 'learning_rate/action_model': 4.2857616100902886e-05, 'learning_rate/action_query_proj': 4.2857616100902886e-05, 'learning_rate/base': 1.0714404025225722e-05, 'epoch': 0.48}) +[2026-06-30 02:31:38,326][starVLA.training.train_starvla][INFO] - Step 1123, Loss: {'train/loss': 3.841379293589853e-05, 'train/grad_norm_pre_clip': 1.0377871990203857, 'timing/data': 0.0003248730208724737, 'timing/model': 1.2302669209893793, 'learning_rate/action_model': 4.278360944696704e-05, 'learning_rate/action_query_proj': 4.278360944696704e-05, 'learning_rate/base': 1.069590236174176e-05, 'epoch': 0.48}) +[2026-06-30 02:32:10,881][starVLA.training.train_starvla][INFO] - Step 1124, Loss: {'train/loss': 0.14370879530906677, 'train/grad_norm_pre_clip': 3.5125110149383545, 'timing/data': 0.00030195899307727814, 'timing/model': 1.222920604981482, 'learning_rate/action_model': 4.270962553355871e-05, 'learning_rate/action_query_proj': 4.270962553355871e-05, 'learning_rate/base': 1.0677406383389677e-05, 'epoch': 0.48}) +[2026-06-30 02:32:43,547][starVLA.training.train_starvla][INFO] - Step 1125, Loss: {'train/loss': 0.004426866769790649, 'train/grad_norm_pre_clip': 1.3703904151916504, 'timing/data': 0.00020308420062065125, 'timing/model': 1.2296060489024967, 'learning_rate/action_model': 4.263566454322585e-05, 'learning_rate/action_query_proj': 4.263566454322585e-05, 'learning_rate/base': 1.0658916135806462e-05, 'epoch': 0.48}) +[2026-06-30 02:33:16,223][starVLA.training.train_starvla][INFO] - Step 1126, Loss: {'train/loss': 5.9360918385209516e-05, 'train/grad_norm_pre_clip': 3.6924450397491455, 'timing/data': 0.0002537388354539871, 'timing/model': 1.2310049799270928, 'learning_rate/action_model': 4.2561726658459846e-05, 'learning_rate/action_query_proj': 4.2561726658459846e-05, 'learning_rate/base': 1.0640431664614961e-05, 'epoch': 0.48}) +[2026-06-30 02:33:48,916][starVLA.training.train_starvla][INFO] - Step 1127, Loss: {'train/loss': 0.46923384070396423, 'train/grad_norm_pre_clip': 1.6107734441757202, 'timing/data': 0.00025110412389039993, 'timing/model': 1.2250210728961974, 'learning_rate/action_model': 4.248781206169507e-05, 'learning_rate/action_query_proj': 4.248781206169507e-05, 'learning_rate/base': 1.0621953015423768e-05, 'epoch': 0.48}) +[2026-06-30 02:34:21,589][starVLA.training.train_starvla][INFO] - Step 1128, Loss: {'train/loss': 3.1261959520634264e-05, 'train/grad_norm_pre_clip': 1.9942541122436523, 'timing/data': 0.0002601600717753172, 'timing/model': 1.2285994738340378, 'learning_rate/action_model': 4.241392093530846e-05, 'learning_rate/action_query_proj': 4.241392093530846e-05, 'learning_rate/base': 1.0603480233827116e-05, 'epoch': 0.49}) +[2026-06-30 02:34:54,213][starVLA.training.train_starvla][INFO] - Step 1129, Loss: {'train/loss': 0.017681652680039406, 'train/grad_norm_pre_clip': 3.1152877807617188, 'timing/data': 0.00026152306236326694, 'timing/model': 1.2262521779630333, 'learning_rate/action_model': 4.234005346161902e-05, 'learning_rate/action_query_proj': 4.234005346161902e-05, 'learning_rate/base': 1.0585013365404755e-05, 'epoch': 0.49}) +[2026-06-30 02:35:26,873][starVLA.training.train_starvla][INFO] - Step 1130, Loss: {'train/loss': 0.015831243246793747, 'train/grad_norm_pre_clip': 2.3672454357147217, 'timing/data': 0.00035773497074842453, 'timing/model': 1.2272825529798865, 'learning_rate/action_model': 4.2266209822887416e-05, 'learning_rate/action_query_proj': 4.2266209822887416e-05, 'learning_rate/base': 1.0566552455721854e-05, 'epoch': 0.49}) +[2026-06-30 02:35:59,525][starVLA.training.train_starvla][INFO] - Step 1131, Loss: {'train/loss': 0.1154996007680893, 'train/grad_norm_pre_clip': 4.092158317565918, 'timing/data': 0.0002664730418473482, 'timing/model': 1.2304608400445431, 'learning_rate/action_model': 4.219239020131543e-05, 'learning_rate/action_query_proj': 4.219239020131543e-05, 'learning_rate/base': 1.0548097550328858e-05, 'epoch': 0.49}) +[2026-06-30 02:36:32,139][starVLA.training.train_starvla][INFO] - Step 1132, Loss: {'train/loss': 8.227863145293668e-05, 'train/grad_norm_pre_clip': 3.277371883392334, 'timing/data': 0.00033488194458186626, 'timing/model': 1.229691389016807, 'learning_rate/action_model': 4.21185947790457e-05, 'learning_rate/action_query_proj': 4.21185947790457e-05, 'learning_rate/base': 1.0529648694761426e-05, 'epoch': 0.49}) +[2026-06-30 02:37:04,808][starVLA.training.train_starvla][INFO] - Step 1133, Loss: {'train/loss': 0.0020823136437684298, 'train/grad_norm_pre_clip': 1.1359566450119019, 'timing/data': 0.00039895297959446907, 'timing/model': 1.2294972850941122, 'learning_rate/action_model': 4.2044823738161055e-05, 'learning_rate/action_query_proj': 4.2044823738161055e-05, 'learning_rate/base': 1.0511205934540264e-05, 'epoch': 0.49}) +[2026-06-30 02:37:37,456][starVLA.training.train_starvla][INFO] - Step 1134, Loss: {'train/loss': 0.02195935882627964, 'train/grad_norm_pre_clip': 1.3311654329299927, 'timing/data': 0.00020635896362364292, 'timing/model': 1.2314832520205528, 'learning_rate/action_model': 4.197107726068425e-05, 'learning_rate/action_query_proj': 4.197107726068425e-05, 'learning_rate/base': 1.0492769315171062e-05, 'epoch': 0.49}) +[2026-06-30 02:38:10,118][starVLA.training.train_starvla][INFO] - Step 1135, Loss: {'train/loss': 0.00037179014179855585, 'train/grad_norm_pre_clip': 1.8825739622116089, 'timing/data': 0.00021135900169610977, 'timing/model': 1.2322874409146607, 'learning_rate/action_model': 4.189735552857734e-05, 'learning_rate/action_query_proj': 4.189735552857734e-05, 'learning_rate/base': 1.0474338882144334e-05, 'epoch': 0.49}) +[2026-06-30 02:38:42,757][starVLA.training.train_starvla][INFO] - Step 1136, Loss: {'train/loss': 0.00011245725909247994, 'train/grad_norm_pre_clip': 2.4551498889923096, 'timing/data': 0.0004775610286742449, 'timing/model': 1.2341224730480462, 'learning_rate/action_model': 4.182365872374139e-05, 'learning_rate/action_query_proj': 4.182365872374139e-05, 'learning_rate/base': 1.0455914680935347e-05, 'epoch': 0.49}) +[2026-06-30 02:39:15,418][starVLA.training.train_starvla][INFO] - Step 1137, Loss: {'train/loss': 0.001465774024836719, 'train/grad_norm_pre_clip': 2.6020026206970215, 'timing/data': 0.00027238414622843266, 'timing/model': 1.2342466479167342, 'learning_rate/action_model': 4.174998702801593e-05, 'learning_rate/action_query_proj': 4.174998702801593e-05, 'learning_rate/base': 1.0437496757003982e-05, 'epoch': 0.49}) +[2026-06-30 02:39:48,074][starVLA.training.train_starvla][INFO] - Step 1138, Loss: {'train/loss': 0.05576840788125992, 'train/grad_norm_pre_clip': 2.601134777069092, 'timing/data': 0.0002136528491973877, 'timing/model': 1.2365168910473585, 'learning_rate/action_model': 4.1676340623178574e-05, 'learning_rate/action_query_proj': 4.1676340623178574e-05, 'learning_rate/base': 1.0419085155794643e-05, 'epoch': 0.49}) +[2026-06-30 02:40:20,702][starVLA.training.train_starvla][INFO] - Step 1139, Loss: {'train/loss': 0.062033556401729584, 'train/grad_norm_pre_clip': 2.918290853500366, 'timing/data': 0.00020910403691232204, 'timing/model': 1.2267406999599189, 'learning_rate/action_model': 4.1602719690944466e-05, 'learning_rate/action_query_proj': 4.1602719690944466e-05, 'learning_rate/base': 1.0400679922736117e-05, 'epoch': 0.49}) +[2026-06-30 02:40:53,380][starVLA.training.train_starvla][INFO] - Step 1140, Loss: {'train/loss': 0.004340282175689936, 'train/grad_norm_pre_clip': 0.9512229561805725, 'timing/data': 0.0002551709767431021, 'timing/model': 1.2330415521282703, 'learning_rate/action_model': 4.152912441296597e-05, 'learning_rate/action_query_proj': 4.152912441296597e-05, 'learning_rate/base': 1.0382281103241492e-05, 'epoch': 0.49}) +[2026-06-30 02:41:26,073][starVLA.training.train_starvla][INFO] - Step 1141, Loss: {'train/loss': 8.881052053766325e-06, 'train/grad_norm_pre_clip': 4.636150360107422, 'timing/data': 0.00023356103338301182, 'timing/model': 1.2323800951708108, 'learning_rate/action_model': 4.1455554970832095e-05, 'learning_rate/action_query_proj': 4.1455554970832095e-05, 'learning_rate/base': 1.0363888742708024e-05, 'epoch': 0.49}) +[2026-06-30 02:41:58,717][starVLA.training.train_starvla][INFO] - Step 1142, Loss: {'train/loss': 0.00019061185594182462, 'train/grad_norm_pre_clip': 1.391892433166504, 'timing/data': 0.0002105669118463993, 'timing/model': 1.2300771560985595, 'learning_rate/action_model': 4.138201154606815e-05, 'learning_rate/action_query_proj': 4.138201154606815e-05, 'learning_rate/base': 1.0345502886517037e-05, 'epoch': 0.49}) +[2026-06-30 02:42:31,535][starVLA.training.train_starvla][INFO] - Step 1143, Loss: {'train/loss': 0.08256691694259644, 'train/grad_norm_pre_clip': 4.430312633514404, 'timing/data': 0.0003180890344083309, 'timing/model': 1.2311945331748575, 'learning_rate/action_model': 4.13084943201352e-05, 'learning_rate/action_query_proj': 4.13084943201352e-05, 'learning_rate/base': 1.03271235800338e-05, 'epoch': 0.49}) +[2026-06-30 02:43:04,224][starVLA.training.train_starvla][INFO] - Step 1144, Loss: {'train/loss': 0.054696571081876755, 'train/grad_norm_pre_clip': 2.9770538806915283, 'timing/data': 0.000254359794780612, 'timing/model': 1.2314184890128672, 'learning_rate/action_model': 4.123500347442971e-05, 'learning_rate/action_query_proj': 4.123500347442971e-05, 'learning_rate/base': 1.0308750868607427e-05, 'epoch': 0.49}) +[2026-06-30 02:43:36,890][starVLA.training.train_starvla][INFO] - Step 1145, Loss: {'train/loss': 0.024227073416113853, 'train/grad_norm_pre_clip': 1.7244491577148438, 'timing/data': 0.00020002713426947594, 'timing/model': 1.2327350459527224, 'learning_rate/action_model': 4.1161539190283027e-05, 'learning_rate/action_query_proj': 4.1161539190283027e-05, 'learning_rate/base': 1.0290384797570757e-05, 'epoch': 0.49}) +[2026-06-30 02:44:09,569][starVLA.training.train_starvla][INFO] - Step 1146, Loss: {'train/loss': 1.1920817087229807e-05, 'train/grad_norm_pre_clip': 6.097003936767578, 'timing/data': 0.0002608019858598709, 'timing/model': 1.2279417398385704, 'learning_rate/action_model': 4.108810164896098e-05, 'learning_rate/action_query_proj': 4.108810164896098e-05, 'learning_rate/base': 1.0272025412240245e-05, 'epoch': 0.49}) +[2026-06-30 02:44:42,237][starVLA.training.train_starvla][INFO] - Step 1147, Loss: {'train/loss': 7.369591185124591e-05, 'train/grad_norm_pre_clip': 2.3453335762023926, 'timing/data': 0.0004703172016888857, 'timing/model': 1.230739715974778, 'learning_rate/action_model': 4.101469103166338e-05, 'learning_rate/action_query_proj': 4.101469103166338e-05, 'learning_rate/base': 1.0253672757915845e-05, 'epoch': 0.49}) +[2026-06-30 02:45:14,921][starVLA.training.train_starvla][INFO] - Step 1148, Loss: {'train/loss': 1.0192325134994462e-05, 'train/grad_norm_pre_clip': 4.2785773277282715, 'timing/data': 0.0002443511039018631, 'timing/model': 1.2332754309754819, 'learning_rate/action_model': 4.094130751952367e-05, 'learning_rate/action_query_proj': 4.094130751952367e-05, 'learning_rate/base': 1.0235326879880917e-05, 'epoch': 0.49}) +[2026-06-30 02:45:47,535][starVLA.training.train_starvla][INFO] - Step 1149, Loss: {'train/loss': 0.07273131608963013, 'train/grad_norm_pre_clip': 2.9570953845977783, 'timing/data': 0.00019380496814846992, 'timing/model': 1.2294616620056331, 'learning_rate/action_model': 4.0867951293608306e-05, 'learning_rate/action_query_proj': 4.0867951293608306e-05, 'learning_rate/base': 1.0216987823402077e-05, 'epoch': 0.49}) +[2026-06-30 02:46:20,169][starVLA.training.train_starvla][INFO] - Step 1150, Loss: {'train/loss': 0.0008712345152162015, 'train/grad_norm_pre_clip': 1.5609619617462158, 'timing/data': 0.00025738589465618134, 'timing/model': 1.2299900210928172, 'learning_rate/action_model': 4.079462253491656e-05, 'learning_rate/action_query_proj': 4.079462253491656e-05, 'learning_rate/base': 1.019865563372914e-05, 'epoch': 0.49}) +[2026-06-30 02:46:52,850][starVLA.training.train_starvla][INFO] - Step 1151, Loss: {'train/loss': 0.011495612561702728, 'train/grad_norm_pre_clip': 3.9916045665740967, 'timing/data': 0.00027020880952477455, 'timing/model': 1.2286615958437324, 'learning_rate/action_model': 4.072132142437977e-05, 'learning_rate/action_query_proj': 4.072132142437977e-05, 'learning_rate/base': 1.0180330356094943e-05, 'epoch': 0.5}) +[2026-06-30 02:47:25,539][starVLA.training.train_starvla][INFO] - Step 1152, Loss: {'train/loss': 0.000878776132594794, 'train/grad_norm_pre_clip': 1.0992804765701294, 'timing/data': 0.0002193329855799675, 'timing/model': 1.2311408640816808, 'learning_rate/action_model': 4.064804814286122e-05, 'learning_rate/action_query_proj': 4.064804814286122e-05, 'learning_rate/base': 1.0162012035715305e-05, 'epoch': 0.5}) +[2026-06-30 02:47:58,247][starVLA.training.train_starvla][INFO] - Step 1153, Loss: {'train/loss': 0.4682646095752716, 'train/grad_norm_pre_clip': 3.3589260578155518, 'timing/data': 0.00019119097851216793, 'timing/model': 1.2340185390785336, 'learning_rate/action_model': 4.05748028711554e-05, 'learning_rate/action_query_proj': 4.05748028711554e-05, 'learning_rate/base': 1.014370071778885e-05, 'epoch': 0.5}) +[2026-06-30 02:48:30,887][starVLA.training.train_starvla][INFO] - Step 1154, Loss: {'train/loss': 0.0001128377698478289, 'train/grad_norm_pre_clip': 1.9064643383026123, 'timing/data': 0.00019844411872327328, 'timing/model': 1.2317248699255288, 'learning_rate/action_model': 4.050158578998776e-05, 'learning_rate/action_query_proj': 4.050158578998776e-05, 'learning_rate/base': 1.012539644749694e-05, 'epoch': 0.5}) +[2026-06-30 02:49:03,537][starVLA.training.train_starvla][INFO] - Step 1155, Loss: {'train/loss': 0.003009149571880698, 'train/grad_norm_pre_clip': 2.6898720264434814, 'timing/data': 0.00019558914937078953, 'timing/model': 1.229736601933837, 'learning_rate/action_model': 4.0428397080014147e-05, 'learning_rate/action_query_proj': 4.0428397080014147e-05, 'learning_rate/base': 1.0107099270003537e-05, 'epoch': 0.5}) +[2026-06-30 02:49:36,148][starVLA.training.train_starvla][INFO] - Step 1156, Loss: {'train/loss': 0.08382651209831238, 'train/grad_norm_pre_clip': 1.9485653638839722, 'timing/data': 0.000287060858681798, 'timing/model': 1.2295653868932277, 'learning_rate/action_model': 4.035523692182045e-05, 'learning_rate/action_query_proj': 4.035523692182045e-05, 'learning_rate/base': 1.0088809230455112e-05, 'epoch': 0.5}) +[2026-06-30 02:50:08,815][starVLA.training.train_starvla][INFO] - Step 1157, Loss: {'train/loss': 0.32898736000061035, 'train/grad_norm_pre_clip': 1.7568814754486084, 'timing/data': 0.00019508693367242813, 'timing/model': 1.2253807990346104, 'learning_rate/action_model': 4.028210549592207e-05, 'learning_rate/action_query_proj': 4.028210549592207e-05, 'learning_rate/base': 1.0070526373980517e-05, 'epoch': 0.5}) +[2026-06-30 02:50:41,535][starVLA.training.train_starvla][INFO] - Step 1158, Loss: {'train/loss': 0.10838530212640762, 'train/grad_norm_pre_clip': 1.5033208131790161, 'timing/data': 0.000289025017991662, 'timing/model': 1.2299291871022433, 'learning_rate/action_model': 4.0209002982763546e-05, 'learning_rate/action_query_proj': 4.0209002982763546e-05, 'learning_rate/base': 1.0052250745690886e-05, 'epoch': 0.5}) +[2026-06-30 02:51:14,216][starVLA.training.train_starvla][INFO] - Step 1159, Loss: {'train/loss': 0.00015034629905130714, 'train/grad_norm_pre_clip': 1.1390372514724731, 'timing/data': 0.00020306301303207874, 'timing/model': 1.2238101449329406, 'learning_rate/action_model': 4.013592956271804e-05, 'learning_rate/action_query_proj': 4.013592956271804e-05, 'learning_rate/base': 1.003398239067951e-05, 'epoch': 0.5}) +[2026-06-30 02:51:46,906][starVLA.training.train_starvla][INFO] - Step 1160, Loss: {'train/loss': 0.1822565495967865, 'train/grad_norm_pre_clip': 3.0669643878936768, 'timing/data': 0.00018942705355584621, 'timing/model': 1.231327675981447, 'learning_rate/action_model': 4.0062885416086976e-05, 'learning_rate/action_query_proj': 4.0062885416086976e-05, 'learning_rate/base': 1.0015721354021744e-05, 'epoch': 0.5}) +[2026-06-30 02:52:19,625][starVLA.training.train_starvla][INFO] - Step 1161, Loss: {'train/loss': 0.682146430015564, 'train/grad_norm_pre_clip': 2.3979032039642334, 'timing/data': 0.00026632193475961685, 'timing/model': 1.231827419018373, 'learning_rate/action_model': 3.9989870723099504e-05, 'learning_rate/action_query_proj': 3.9989870723099504e-05, 'learning_rate/base': 9.997467680774876e-06, 'epoch': 0.5}) +[2026-06-30 02:52:52,279][starVLA.training.train_starvla][INFO] - Step 1162, Loss: {'train/loss': 0.003087032353505492, 'train/grad_norm_pre_clip': 1.8546817302703857, 'timing/data': 0.0003982309717684984, 'timing/model': 1.2286608470603824, 'learning_rate/action_model': 3.991688566391214e-05, 'learning_rate/action_query_proj': 3.991688566391214e-05, 'learning_rate/base': 9.979221415978035e-06, 'epoch': 0.5}) +[2026-06-30 02:53:24,957][starVLA.training.train_starvla][INFO] - Step 1163, Loss: {'train/loss': 0.02585556171834469, 'train/grad_norm_pre_clip': 3.2240025997161865, 'timing/data': 0.0002873418852686882, 'timing/model': 1.228125634137541, 'learning_rate/action_model': 3.984393041860825e-05, 'learning_rate/action_query_proj': 3.984393041860825e-05, 'learning_rate/base': 9.960982604652063e-06, 'epoch': 0.5}) +[2026-06-30 02:53:57,681][starVLA.training.train_starvla][INFO] - Step 1164, Loss: {'train/loss': 0.06437907367944717, 'train/grad_norm_pre_clip': 2.2999267578125, 'timing/data': 0.00028976588509976864, 'timing/model': 1.2332440209574997, 'learning_rate/action_model': 3.977100516719766e-05, 'learning_rate/action_query_proj': 3.977100516719766e-05, 'learning_rate/base': 9.942751291799415e-06, 'epoch': 0.5}) +[2026-06-30 02:54:30,325][starVLA.training.train_starvla][INFO] - Step 1165, Loss: {'train/loss': 0.19100399315357208, 'train/grad_norm_pre_clip': 2.4770240783691406, 'timing/data': 0.0002024918794631958, 'timing/model': 1.2394954531919211, 'learning_rate/action_model': 3.969811008961616e-05, 'learning_rate/action_query_proj': 3.969811008961616e-05, 'learning_rate/base': 9.92452752240404e-06, 'epoch': 0.5}) +[2026-06-30 02:55:03,016][starVLA.training.train_starvla][INFO] - Step 1166, Loss: {'train/loss': 0.004951044917106628, 'train/grad_norm_pre_clip': 2.1002416610717773, 'timing/data': 0.00040209805592894554, 'timing/model': 1.2347537211608142, 'learning_rate/action_model': 3.962524536572513e-05, 'learning_rate/action_query_proj': 3.962524536572513e-05, 'learning_rate/base': 9.906311341431282e-06, 'epoch': 0.5}) +[2026-06-30 02:55:35,695][starVLA.training.train_starvla][INFO] - Step 1167, Loss: {'train/loss': 0.06349485367536545, 'train/grad_norm_pre_clip': 2.2397289276123047, 'timing/data': 0.00019279285334050655, 'timing/model': 1.2316271220333874, 'learning_rate/action_model': 3.955241117531101e-05, 'learning_rate/action_query_proj': 3.955241117531101e-05, 'learning_rate/base': 9.888102793827753e-06, 'epoch': 0.5}) +[2026-06-30 02:56:08,348][starVLA.training.train_starvla][INFO] - Step 1168, Loss: {'train/loss': 8.672402145748492e-06, 'train/grad_norm_pre_clip': 1.6384690999984741, 'timing/data': 0.0001903290394693613, 'timing/model': 1.2268917090259492, 'learning_rate/action_model': 3.947960769808496e-05, 'learning_rate/action_query_proj': 3.947960769808496e-05, 'learning_rate/base': 9.86990192452124e-06, 'epoch': 0.5}) +[2026-06-30 02:56:41,002][starVLA.training.train_starvla][INFO] - Step 1169, Loss: {'train/loss': 9.298226359533146e-06, 'train/grad_norm_pre_clip': 1.7079589366912842, 'timing/data': 0.00019216304644942284, 'timing/model': 1.2316586340311915, 'learning_rate/action_model': 3.940683511368227e-05, 'learning_rate/action_query_proj': 3.940683511368227e-05, 'learning_rate/base': 9.851708778420567e-06, 'epoch': 0.5}) +[2026-06-30 02:57:13,673][starVLA.training.train_starvla][INFO] - Step 1170, Loss: {'train/loss': 0.07555381208658218, 'train/grad_norm_pre_clip': 1.295296311378479, 'timing/data': 0.00032655615359544754, 'timing/model': 1.229352978989482, 'learning_rate/action_model': 3.933409360166211e-05, 'learning_rate/action_query_proj': 3.933409360166211e-05, 'learning_rate/base': 9.833523400415528e-06, 'epoch': 0.5}) +[2026-06-30 02:57:46,354][starVLA.training.train_starvla][INFO] - Step 1171, Loss: {'train/loss': 0.1040281355381012, 'train/grad_norm_pre_clip': 2.768714427947998, 'timing/data': 0.00020630890503525734, 'timing/model': 1.2318782941438258, 'learning_rate/action_model': 3.926138334150688e-05, 'learning_rate/action_query_proj': 3.926138334150688e-05, 'learning_rate/base': 9.81534583537672e-06, 'epoch': 0.5}) +[2026-06-30 02:58:18,986][starVLA.training.train_starvla][INFO] - Step 1172, Loss: {'train/loss': 0.16285499930381775, 'train/grad_norm_pre_clip': 1.2966996431350708, 'timing/data': 0.00021974393166601658, 'timing/model': 1.23184061399661, 'learning_rate/action_model': 3.918870451262198e-05, 'learning_rate/action_query_proj': 3.918870451262198e-05, 'learning_rate/base': 9.797176128155495e-06, 'epoch': 0.5}) +[2026-06-30 02:58:51,652][starVLA.training.train_starvla][INFO] - Step 1173, Loss: {'train/loss': 0.0003617858455982059, 'train/grad_norm_pre_clip': 1.9861557483673096, 'timing/data': 0.00020693987607955933, 'timing/model': 1.237989128800109, 'learning_rate/action_model': 3.9116057294335146e-05, 'learning_rate/action_query_proj': 3.9116057294335146e-05, 'learning_rate/base': 9.779014323583786e-06, 'epoch': 0.5}) +[2026-06-30 02:59:24,278][starVLA.training.train_starvla][INFO] - Step 1174, Loss: {'train/loss': 5.569468703470193e-05, 'train/grad_norm_pre_clip': 2.6692118644714355, 'timing/data': 0.0002063489519059658, 'timing/model': 1.233290976146236, 'learning_rate/action_model': 3.9043441865896186e-05, 'learning_rate/action_query_proj': 3.9043441865896186e-05, 'learning_rate/base': 9.760860466474047e-06, 'epoch': 0.51}) +[2026-06-30 02:59:56,951][starVLA.training.train_starvla][INFO] - Step 1175, Loss: {'train/loss': 0.00015802253619767725, 'train/grad_norm_pre_clip': 2.105229139328003, 'timing/data': 0.00020841299556195736, 'timing/model': 1.2349358049686998, 'learning_rate/action_model': 3.8970858406476436e-05, 'learning_rate/action_query_proj': 3.8970858406476436e-05, 'learning_rate/base': 9.742714601619109e-06, 'epoch': 0.51}) +[2026-06-30 03:00:29,629][starVLA.training.train_starvla][INFO] - Step 1176, Loss: {'train/loss': 0.2535322904586792, 'train/grad_norm_pre_clip': 3.6886539459228516, 'timing/data': 0.0002084840089082718, 'timing/model': 1.2283080308698118, 'learning_rate/action_model': 3.889830709516839e-05, 'learning_rate/action_query_proj': 3.889830709516839e-05, 'learning_rate/base': 9.724576773792098e-06, 'epoch': 0.51}) +[2026-06-30 03:01:02,269][starVLA.training.train_starvla][INFO] - Step 1177, Loss: {'train/loss': 0.00019200460519641638, 'train/grad_norm_pre_clip': 2.4240498542785645, 'timing/data': 0.00020621903240680695, 'timing/model': 1.2324417030904442, 'learning_rate/action_model': 3.8825788110985186e-05, 'learning_rate/action_query_proj': 3.8825788110985186e-05, 'learning_rate/base': 9.706447027746297e-06, 'epoch': 0.51}) +[2026-06-30 03:01:34,919][starVLA.training.train_starvla][INFO] - Step 1178, Loss: {'train/loss': 2.205368673457997e-06, 'train/grad_norm_pre_clip': 3.0689778327941895, 'timing/data': 0.0003725930582731962, 'timing/model': 1.2323484800290316, 'learning_rate/action_model': 3.875330163286021e-05, 'learning_rate/action_query_proj': 3.875330163286021e-05, 'learning_rate/base': 9.688325408215052e-06, 'epoch': 0.51}) +[2026-06-30 03:02:07,553][starVLA.training.train_starvla][INFO] - Step 1179, Loss: {'train/loss': 0.00011561022256501019, 'train/grad_norm_pre_clip': 1.680972695350647, 'timing/data': 0.00038773100823163986, 'timing/model': 1.2272451978642493, 'learning_rate/action_model': 3.8680847839646636e-05, 'learning_rate/action_query_proj': 3.8680847839646636e-05, 'learning_rate/base': 9.670211959911659e-06, 'epoch': 0.51}) +[2026-06-30 03:02:40,184][starVLA.training.train_starvla][INFO] - Step 1180, Loss: {'train/loss': 0.0645981952548027, 'train/grad_norm_pre_clip': 1.6617342233657837, 'timing/data': 0.00022071599960327148, 'timing/model': 1.2277374600525945, 'learning_rate/action_model': 3.8608426910117e-05, 'learning_rate/action_query_proj': 3.8608426910117e-05, 'learning_rate/base': 9.65210672752925e-06, 'epoch': 0.51}) +[2026-06-30 03:03:12,850][starVLA.training.train_starvla][INFO] - Step 1181, Loss: {'train/loss': 0.011151024140417576, 'train/grad_norm_pre_clip': 1.2102150917053223, 'timing/data': 0.0002598708961158991, 'timing/model': 1.2325411608908325, 'learning_rate/action_model': 3.8536039022962736e-05, 'learning_rate/action_query_proj': 3.8536039022962736e-05, 'learning_rate/base': 9.634009755740684e-06, 'epoch': 0.51}) +[2026-06-30 03:03:45,428][starVLA.training.train_starvla][INFO] - Step 1182, Loss: {'train/loss': 1.817939619286335e-06, 'train/grad_norm_pre_clip': 2.693132162094116, 'timing/data': 0.00019726203754544258, 'timing/model': 1.2315672361291945, 'learning_rate/action_model': 3.846368435679379e-05, 'learning_rate/action_query_proj': 3.846368435679379e-05, 'learning_rate/base': 9.615921089198447e-06, 'epoch': 0.51}) +[2026-06-30 03:04:18,076][starVLA.training.train_starvla][INFO] - Step 1183, Loss: {'train/loss': 3.933891093765851e-06, 'train/grad_norm_pre_clip': 2.7181713581085205, 'timing/data': 0.00020712101832032204, 'timing/model': 1.2337424890138209, 'learning_rate/action_model': 3.839136309013808e-05, 'learning_rate/action_query_proj': 3.839136309013808e-05, 'learning_rate/base': 9.59784077253452e-06, 'epoch': 0.51}) +[2026-06-30 03:04:50,733][starVLA.training.train_starvla][INFO] - Step 1184, Loss: {'train/loss': 0.00013664069410879165, 'train/grad_norm_pre_clip': 3.6582329273223877, 'timing/data': 0.00029598898254334927, 'timing/model': 1.2306031270418316, 'learning_rate/action_model': 3.8319075401441144e-05, 'learning_rate/action_query_proj': 3.8319075401441144e-05, 'learning_rate/base': 9.579768850360286e-06, 'epoch': 0.51}) +[2026-06-30 03:05:23,374][starVLA.training.train_starvla][INFO] - Step 1185, Loss: {'train/loss': 0.013671919703483582, 'train/grad_norm_pre_clip': 4.1562676429748535, 'timing/data': 0.00025511113926768303, 'timing/model': 1.2301637770142406, 'learning_rate/action_model': 3.8246821469065676e-05, 'learning_rate/action_query_proj': 3.8246821469065676e-05, 'learning_rate/base': 9.561705367266419e-06, 'epoch': 0.51}) +[2026-06-30 03:05:56,029][starVLA.training.train_starvla][INFO] - Step 1186, Loss: {'train/loss': 0.4416464865207672, 'train/grad_norm_pre_clip': 1.5331521034240723, 'timing/data': 0.0002369871363043785, 'timing/model': 1.2316324959974736, 'learning_rate/action_model': 3.817460147129109e-05, 'learning_rate/action_query_proj': 3.817460147129109e-05, 'learning_rate/base': 9.543650367822773e-06, 'epoch': 0.51}) +[2026-06-30 03:06:28,622][starVLA.training.train_starvla][INFO] - Step 1187, Loss: {'train/loss': 0.000778637477196753, 'train/grad_norm_pre_clip': 1.978398084640503, 'timing/data': 0.00019738194532692432, 'timing/model': 1.22708023712039, 'learning_rate/action_model': 3.8102415586313e-05, 'learning_rate/action_query_proj': 3.8102415586313e-05, 'learning_rate/base': 9.52560389657825e-06, 'epoch': 0.51}) +[2026-06-30 03:07:01,256][starVLA.training.train_starvla][INFO] - Step 1188, Loss: {'train/loss': 0.011760813184082508, 'train/grad_norm_pre_clip': 2.164348602294922, 'timing/data': 0.0002065589651465416, 'timing/model': 1.2337777181528509, 'learning_rate/action_model': 3.803026399224296e-05, 'learning_rate/action_query_proj': 3.803026399224296e-05, 'learning_rate/base': 9.50756599806074e-06, 'epoch': 0.51}) +[2026-06-30 03:07:33,924][starVLA.training.train_starvla][INFO] - Step 1189, Loss: {'train/loss': 0.01633579097688198, 'train/grad_norm_pre_clip': 2.7544214725494385, 'timing/data': 0.00019850395619869232, 'timing/model': 1.2307227218989283, 'learning_rate/action_model': 3.79581468671078e-05, 'learning_rate/action_query_proj': 3.79581468671078e-05, 'learning_rate/base': 9.48953671677695e-06, 'epoch': 0.51}) +[2026-06-30 03:08:06,634][starVLA.training.train_starvla][INFO] - Step 1190, Loss: {'train/loss': 0.329004168510437, 'train/grad_norm_pre_clip': 2.206989288330078, 'timing/data': 0.000418288866057992, 'timing/model': 1.2386034841183573, 'learning_rate/action_model': 3.788606438884941e-05, 'learning_rate/action_query_proj': 3.788606438884941e-05, 'learning_rate/base': 9.471516097212353e-06, 'epoch': 0.51}) +[2026-06-30 03:08:39,311][starVLA.training.train_starvla][INFO] - Step 1191, Loss: {'train/loss': 1.460312660128693e-06, 'train/grad_norm_pre_clip': 2.327221632003784, 'timing/data': 0.00019939709454774857, 'timing/model': 1.2287839080672711, 'learning_rate/action_model': 3.781401673532411e-05, 'learning_rate/action_query_proj': 3.781401673532411e-05, 'learning_rate/base': 9.453504183831028e-06, 'epoch': 0.51}) +[2026-06-30 03:09:11,956][starVLA.training.train_starvla][INFO] - Step 1192, Loss: {'train/loss': 0.0012457265984266996, 'train/grad_norm_pre_clip': 2.5692267417907715, 'timing/data': 0.0003662200178951025, 'timing/model': 1.2284525791183114, 'learning_rate/action_model': 3.7742004084302326e-05, 'learning_rate/action_query_proj': 3.7742004084302326e-05, 'learning_rate/base': 9.435501021075582e-06, 'epoch': 0.51}) +[2026-06-30 03:09:44,608][starVLA.training.train_starvla][INFO] - Step 1193, Loss: {'train/loss': 0.0029273100662976503, 'train/grad_norm_pre_clip': 4.142216682434082, 'timing/data': 0.00025009200908243656, 'timing/model': 1.2324002699460834, 'learning_rate/action_model': 3.767002661346812e-05, 'learning_rate/action_query_proj': 3.767002661346812e-05, 'learning_rate/base': 9.41750665336703e-06, 'epoch': 0.51}) +[2026-06-30 03:10:17,262][starVLA.training.train_starvla][INFO] - Step 1194, Loss: {'train/loss': 0.002495564054697752, 'train/grad_norm_pre_clip': 4.076322078704834, 'timing/data': 0.0002911798655986786, 'timing/model': 1.2298047789372504, 'learning_rate/action_model': 3.759808450041875e-05, 'learning_rate/action_query_proj': 3.759808450041875e-05, 'learning_rate/base': 9.399521125104687e-06, 'epoch': 0.51}) +[2026-06-30 03:10:50,109][starVLA.training.train_starvla][INFO] - Step 1195, Loss: {'train/loss': 0.10257064551115036, 'train/grad_norm_pre_clip': 1.9332002401351929, 'timing/data': 0.00020862300880253315, 'timing/model': 1.2316384220030159, 'learning_rate/action_model': 3.7526177922664204e-05, 'learning_rate/action_query_proj': 3.7526177922664204e-05, 'learning_rate/base': 9.381544480666051e-06, 'epoch': 0.51}) +[2026-06-30 03:11:22,710][starVLA.training.train_starvla][INFO] - Step 1196, Loss: {'train/loss': 0.13356034457683563, 'train/grad_norm_pre_clip': 2.398531913757324, 'timing/data': 0.0002097447868436575, 'timing/model': 1.2182224309071898, 'learning_rate/action_model': 3.7454307057626854e-05, 'learning_rate/action_query_proj': 3.7454307057626854e-05, 'learning_rate/base': 9.363576764406713e-06, 'epoch': 0.51}) +[2026-06-30 03:11:55,356][starVLA.training.train_starvla][INFO] - Step 1197, Loss: {'train/loss': 0.01202476304024458, 'train/grad_norm_pre_clip': 3.666034460067749, 'timing/data': 0.0001992359757423401, 'timing/model': 1.2298365531023592, 'learning_rate/action_model': 3.738247208264088e-05, 'learning_rate/action_query_proj': 3.738247208264088e-05, 'learning_rate/base': 9.34561802066022e-06, 'epoch': 0.52}) +[2026-06-30 03:12:28,019][starVLA.training.train_starvla][INFO] - Step 1198, Loss: {'train/loss': 0.021567145362496376, 'train/grad_norm_pre_clip': 2.041388750076294, 'timing/data': 0.0003129600081592798, 'timing/model': 1.227196505991742, 'learning_rate/action_model': 3.7310673174951966e-05, 'learning_rate/action_query_proj': 3.7310673174951966e-05, 'learning_rate/base': 9.327668293737992e-06, 'epoch': 0.52}) +[2026-06-30 03:13:00,670][starVLA.training.train_starvla][INFO] - Step 1199, Loss: {'train/loss': 0.00036563989124260843, 'train/grad_norm_pre_clip': 3.0466256141662598, 'timing/data': 0.0001953889150172472, 'timing/model': 1.2344314269721508, 'learning_rate/action_model': 3.723891051171677e-05, 'learning_rate/action_query_proj': 3.723891051171677e-05, 'learning_rate/base': 9.309727627929192e-06, 'epoch': 0.52}) +[2026-06-30 03:13:33,291][starVLA.training.train_starvla][INFO] - Step 1200, Loss: {'train/loss': 0.0053076911717653275, 'train/grad_norm_pre_clip': 1.9668893814086914, 'timing/data': 0.0002996551338583231, 'timing/model': 1.2260166280902922, 'learning_rate/action_model': 3.716718427000253e-05, 'learning_rate/action_query_proj': 3.716718427000253e-05, 'learning_rate/base': 9.291796067500632e-06, 'epoch': 0.52}) +[2026-06-30 03:14:05,928][starVLA.training.train_starvla][INFO] - Step 1201, Loss: {'train/loss': 2.255985964438878e-05, 'train/grad_norm_pre_clip': 2.149780035018921, 'timing/data': 0.0002723040524870157, 'timing/model': 1.2273986290674657, 'learning_rate/action_model': 3.70954946267866e-05, 'learning_rate/action_query_proj': 3.70954946267866e-05, 'learning_rate/base': 9.27387365669665e-06, 'epoch': 0.52}) +[2026-06-30 03:14:38,596][starVLA.training.train_starvla][INFO] - Step 1202, Loss: {'train/loss': 0.0008378066122531891, 'train/grad_norm_pre_clip': 3.581737995147705, 'timing/data': 0.0003406028263270855, 'timing/model': 1.2311570309102535, 'learning_rate/action_model': 3.702384175895609e-05, 'learning_rate/action_query_proj': 3.702384175895609e-05, 'learning_rate/base': 9.255960439739023e-06, 'epoch': 0.52}) +[2026-06-30 03:15:11,253][starVLA.training.train_starvla][INFO] - Step 1203, Loss: {'train/loss': 0.0062668281607329845, 'train/grad_norm_pre_clip': 0.9788334965705872, 'timing/data': 0.00028525805100798607, 'timing/model': 1.2315852909814566, 'learning_rate/action_model': 3.6952225843307286e-05, 'learning_rate/action_query_proj': 3.6952225843307286e-05, 'learning_rate/base': 9.238056460826822e-06, 'epoch': 0.52}) +[2026-06-30 03:15:43,870][starVLA.training.train_starvla][INFO] - Step 1204, Loss: {'train/loss': 0.06275799870491028, 'train/grad_norm_pre_clip': 2.1443443298339844, 'timing/data': 0.00019338494166731834, 'timing/model': 1.228827232029289, 'learning_rate/action_model': 3.688064705654538e-05, 'learning_rate/action_query_proj': 3.688064705654538e-05, 'learning_rate/base': 9.220161764136345e-06, 'epoch': 0.52}) +[2026-06-30 03:16:16,525][starVLA.training.train_starvla][INFO] - Step 1205, Loss: {'train/loss': 0.00017124321311712265, 'train/grad_norm_pre_clip': 2.7970569133758545, 'timing/data': 0.0002733150031417608, 'timing/model': 1.2180656150449067, 'learning_rate/action_model': 3.6809105575283866e-05, 'learning_rate/action_query_proj': 3.6809105575283866e-05, 'learning_rate/base': 9.202276393820967e-06, 'epoch': 0.52}) +[2026-06-30 03:16:49,181][starVLA.training.train_starvla][INFO] - Step 1206, Loss: {'train/loss': 0.0031519467011094093, 'train/grad_norm_pre_clip': 2.281395673751831, 'timing/data': 0.00019752210937440395, 'timing/model': 1.2340222250204533, 'learning_rate/action_model': 3.6737601576044286e-05, 'learning_rate/action_query_proj': 3.6737601576044286e-05, 'learning_rate/base': 9.184400394011072e-06, 'epoch': 0.52}) +[2026-06-30 03:17:21,848][starVLA.training.train_starvla][INFO] - Step 1207, Loss: {'train/loss': 0.030635425820946693, 'train/grad_norm_pre_clip': 5.2605156898498535, 'timing/data': 0.000194967957213521, 'timing/model': 1.2279572780244052, 'learning_rate/action_model': 3.66661352352556e-05, 'learning_rate/action_query_proj': 3.66661352352556e-05, 'learning_rate/base': 9.1665338088139e-06, 'epoch': 0.52}) +[2026-06-30 03:17:54,496][starVLA.training.train_starvla][INFO] - Step 1208, Loss: {'train/loss': 0.17175446450710297, 'train/grad_norm_pre_clip': 4.119332790374756, 'timing/data': 0.00024323887191712856, 'timing/model': 1.2284033419564366, 'learning_rate/action_model': 3.6594706729253945e-05, 'learning_rate/action_query_proj': 3.6594706729253945e-05, 'learning_rate/base': 9.148676682313486e-06, 'epoch': 0.52}) +[2026-06-30 03:18:27,150][starVLA.training.train_starvla][INFO] - Step 1209, Loss: {'train/loss': 0.02746744081377983, 'train/grad_norm_pre_clip': 2.9200620651245117, 'timing/data': 0.00028037908487021923, 'timing/model': 1.2324720171745867, 'learning_rate/action_model': 3.6523316234282036e-05, 'learning_rate/action_query_proj': 3.6523316234282036e-05, 'learning_rate/base': 9.130829058570509e-06, 'epoch': 0.52}) +[2026-06-30 03:18:59,821][starVLA.training.train_starvla][INFO] - Step 1210, Loss: {'train/loss': 0.00020903108816128224, 'train/grad_norm_pre_clip': 2.1715171337127686, 'timing/data': 0.00021582795307040215, 'timing/model': 1.2328319479711354, 'learning_rate/action_model': 3.6451963926488834e-05, 'learning_rate/action_query_proj': 3.6451963926488834e-05, 'learning_rate/base': 9.112990981622208e-06, 'epoch': 0.52}) +[2026-06-30 03:19:32,478][starVLA.training.train_starvla][INFO] - Step 1211, Loss: {'train/loss': 1.972869358723983e-05, 'train/grad_norm_pre_clip': 2.7704918384552, 'timing/data': 0.00025845784693956375, 'timing/model': 1.2309363551903516, 'learning_rate/action_model': 3.6380649981929045e-05, 'learning_rate/action_query_proj': 3.6380649981929045e-05, 'learning_rate/base': 9.095162495482261e-06, 'epoch': 0.52}) +[2026-06-30 03:20:05,086][starVLA.training.train_starvla][INFO] - Step 1212, Loss: {'train/loss': 0.2991876006126404, 'train/grad_norm_pre_clip': 1.4786226749420166, 'timing/data': 0.00020095892250537872, 'timing/model': 1.2334915359970182, 'learning_rate/action_model': 3.630937457656277e-05, 'learning_rate/action_query_proj': 3.630937457656277e-05, 'learning_rate/base': 9.077343644140692e-06, 'epoch': 0.52}) +[2026-06-30 03:20:37,761][starVLA.training.train_starvla][INFO] - Step 1213, Loss: {'train/loss': 0.0007830674876458943, 'train/grad_norm_pre_clip': 1.4015768766403198, 'timing/data': 0.00025324802845716476, 'timing/model': 1.2350836810655892, 'learning_rate/action_model': 3.6238137886254946e-05, 'learning_rate/action_query_proj': 3.6238137886254946e-05, 'learning_rate/base': 9.059534471563736e-06, 'epoch': 0.52}) +[2026-06-30 03:21:10,392][starVLA.training.train_starvla][INFO] - Step 1214, Loss: {'train/loss': 3.34070464305114e-05, 'train/grad_norm_pre_clip': 2.763193368911743, 'timing/data': 0.00024237693287432194, 'timing/model': 1.2304255550261587, 'learning_rate/action_model': 3.6166940086775067e-05, 'learning_rate/action_query_proj': 3.6166940086775067e-05, 'learning_rate/base': 9.041735021693767e-06, 'epoch': 0.52}) +[2026-06-30 03:21:43,043][starVLA.training.train_starvla][INFO] - Step 1215, Loss: {'train/loss': 1.2874440471932758e-05, 'train/grad_norm_pre_clip': 2.388435125350952, 'timing/data': 0.00020943512208759785, 'timing/model': 1.2315992610529065, 'learning_rate/action_model': 3.6095781353796584e-05, 'learning_rate/action_query_proj': 3.6095781353796584e-05, 'learning_rate/base': 9.023945338449146e-06, 'epoch': 0.52}) +[2026-06-30 03:22:15,700][starVLA.training.train_starvla][INFO] - Step 1216, Loss: {'train/loss': 0.06136605143547058, 'train/grad_norm_pre_clip': 1.3425697088241577, 'timing/data': 0.00024734693579375744, 'timing/model': 1.2292094151489437, 'learning_rate/action_model': 3.602466186289664e-05, 'learning_rate/action_query_proj': 3.602466186289664e-05, 'learning_rate/base': 9.00616546572416e-06, 'epoch': 0.52}) +[2026-06-30 03:22:48,355][starVLA.training.train_starvla][INFO] - Step 1217, Loss: {'train/loss': 1.1750874519348145, 'train/grad_norm_pre_clip': 1.9534144401550293, 'timing/data': 0.0002841257955878973, 'timing/model': 1.2352453209459782, 'learning_rate/action_model': 3.595358178955548e-05, 'learning_rate/action_query_proj': 3.595358178955548e-05, 'learning_rate/base': 8.98839544738887e-06, 'epoch': 0.52}) +[2026-06-30 03:23:21,018][starVLA.training.train_starvla][INFO] - Step 1218, Loss: {'train/loss': 0.036750368773937225, 'train/grad_norm_pre_clip': 1.5580044984817505, 'timing/data': 0.000199456000700593, 'timing/model': 1.2307554630097002, 'learning_rate/action_model': 3.588254130915613e-05, 'learning_rate/action_query_proj': 3.588254130915613e-05, 'learning_rate/base': 8.970635327289032e-06, 'epoch': 0.52}) +[2026-06-30 03:23:53,646][starVLA.training.train_starvla][INFO] - Step 1219, Loss: {'train/loss': 1.6450650946353562e-05, 'train/grad_norm_pre_clip': 1.0135167837142944, 'timing/data': 0.00019544805400073528, 'timing/model': 1.2270408070180565, 'learning_rate/action_model': 3.58115405969839e-05, 'learning_rate/action_query_proj': 3.58115405969839e-05, 'learning_rate/base': 8.952885149245975e-06, 'epoch': 0.52}) +[2026-06-30 03:24:26,347][starVLA.training.train_starvla][INFO] - Step 1220, Loss: {'train/loss': 5.468361632665619e-05, 'train/grad_norm_pre_clip': 0.9787000417709351, 'timing/data': 0.00025921803899109364, 'timing/model': 1.2318086510058492, 'learning_rate/action_model': 3.5740579828226014e-05, 'learning_rate/action_query_proj': 3.5740579828226014e-05, 'learning_rate/base': 8.935144957056504e-06, 'epoch': 0.52}) +[2026-06-30 03:24:59,021][starVLA.training.train_starvla][INFO] - Step 1221, Loss: {'train/loss': 0.00473934318870306, 'train/grad_norm_pre_clip': 2.223606586456299, 'timing/data': 0.00026203482411801815, 'timing/model': 1.2312669409438968, 'learning_rate/action_model': 3.5669659177971085e-05, 'learning_rate/action_query_proj': 3.5669659177971085e-05, 'learning_rate/base': 8.917414794492771e-06, 'epoch': 0.53}) +[2026-06-30 03:25:31,733][starVLA.training.train_starvla][INFO] - Step 1222, Loss: {'train/loss': 0.09129203855991364, 'train/grad_norm_pre_clip': 2.31126070022583, 'timing/data': 0.0002560329157859087, 'timing/model': 1.2320336170960218, 'learning_rate/action_model': 3.55987788212088e-05, 'learning_rate/action_query_proj': 3.55987788212088e-05, 'learning_rate/base': 8.8996947053022e-06, 'epoch': 0.53}) +[2026-06-30 03:26:04,405][starVLA.training.train_starvla][INFO] - Step 1223, Loss: {'train/loss': 0.048400964587926865, 'train/grad_norm_pre_clip': 1.6027255058288574, 'timing/data': 0.00021428405307233334, 'timing/model': 1.2371295299381018, 'learning_rate/action_model': 3.552793893282936e-05, 'learning_rate/action_query_proj': 3.552793893282936e-05, 'learning_rate/base': 8.88198473320734e-06, 'epoch': 0.53}) +[2026-06-30 03:26:37,044][starVLA.training.train_starvla][INFO] - Step 1224, Loss: {'train/loss': 0.1502629816532135, 'train/grad_norm_pre_clip': 2.0296359062194824, 'timing/data': 0.0002460039686411619, 'timing/model': 1.2315817570779473, 'learning_rate/action_model': 3.5457139687623186e-05, 'learning_rate/action_query_proj': 3.5457139687623186e-05, 'learning_rate/base': 8.864284921905797e-06, 'epoch': 0.53}) +[2026-06-30 03:27:09,723][starVLA.training.train_starvla][INFO] - Step 1225, Loss: {'train/loss': 0.00391432736068964, 'train/grad_norm_pre_clip': 1.3351619243621826, 'timing/data': 0.0002048059832304716, 'timing/model': 1.2284948350861669, 'learning_rate/action_model': 3.538638126028033e-05, 'learning_rate/action_query_proj': 3.538638126028033e-05, 'learning_rate/base': 8.846595315070082e-06, 'epoch': 0.53}) +[2026-06-30 03:27:42,365][starVLA.training.train_starvla][INFO] - Step 1226, Loss: {'train/loss': 0.11796648055315018, 'train/grad_norm_pre_clip': 2.0683915615081787, 'timing/data': 0.0002888049930334091, 'timing/model': 1.2400624961592257, 'learning_rate/action_model': 3.531566382539023e-05, 'learning_rate/action_query_proj': 3.531566382539023e-05, 'learning_rate/base': 8.828915956347557e-06, 'epoch': 0.53}) +[2026-06-30 03:28:15,046][starVLA.training.train_starvla][INFO] - Step 1227, Loss: {'train/loss': 0.03069133311510086, 'train/grad_norm_pre_clip': 2.093494415283203, 'timing/data': 0.0002089240588247776, 'timing/model': 1.2303928460460156, 'learning_rate/action_model': 3.5244987557441066e-05, 'learning_rate/action_query_proj': 3.5244987557441066e-05, 'learning_rate/base': 8.811246889360267e-06, 'epoch': 0.53}) +[2026-06-30 03:28:47,715][starVLA.training.train_starvla][INFO] - Step 1228, Loss: {'train/loss': 1.8179189282818697e-05, 'train/grad_norm_pre_clip': 1.8960949182510376, 'timing/data': 0.0001959300134330988, 'timing/model': 1.2317193788476288, 'learning_rate/action_model': 3.5174352630819577e-05, 'learning_rate/action_query_proj': 3.5174352630819577e-05, 'learning_rate/base': 8.793588157704894e-06, 'epoch': 0.53}) +[2026-06-30 03:29:20,413][starVLA.training.train_starvla][INFO] - Step 1229, Loss: {'train/loss': 3.150001430185512e-05, 'train/grad_norm_pre_clip': 3.5321621894836426, 'timing/data': 0.0001940261572599411, 'timing/model': 1.2287421240471303, 'learning_rate/action_model': 3.510375921981035e-05, 'learning_rate/action_query_proj': 3.510375921981035e-05, 'learning_rate/base': 8.775939804952588e-06, 'epoch': 0.53}) +[2026-06-30 03:29:53,041][starVLA.training.train_starvla][INFO] - Step 1230, Loss: {'train/loss': 2.5033909878402483e-06, 'train/grad_norm_pre_clip': 2.4881510734558105, 'timing/data': 0.00020526698790490627, 'timing/model': 1.2256133519113064, 'learning_rate/action_model': 3.5033207498595666e-05, 'learning_rate/action_query_proj': 3.5033207498595666e-05, 'learning_rate/base': 8.758301874648916e-06, 'epoch': 0.53}) +[2026-06-30 03:30:25,645][starVLA.training.train_starvla][INFO] - Step 1231, Loss: {'train/loss': 0.0011578729609027505, 'train/grad_norm_pre_clip': 2.0793941020965576, 'timing/data': 0.00032629515044391155, 'timing/model': 1.2273225460667163, 'learning_rate/action_model': 3.496269764125484e-05, 'learning_rate/action_query_proj': 3.496269764125484e-05, 'learning_rate/base': 8.74067441031371e-06, 'epoch': 0.53}) +[2026-06-30 03:30:58,305][starVLA.training.train_starvla][INFO] - Step 1232, Loss: {'train/loss': 3.2186424050451024e-06, 'train/grad_norm_pre_clip': 2.6049411296844482, 'timing/data': 0.00021245982497930527, 'timing/model': 1.2282337660435587, 'learning_rate/action_model': 3.4892229821763966e-05, 'learning_rate/action_query_proj': 3.4892229821763966e-05, 'learning_rate/base': 8.723057455440992e-06, 'epoch': 0.53}) +[2026-06-30 03:31:30,896][starVLA.training.train_starvla][INFO] - Step 1233, Loss: {'train/loss': 0.1172274723649025, 'train/grad_norm_pre_clip': 2.7097604274749756, 'timing/data': 0.00030364212580025196, 'timing/model': 1.228964684996754, 'learning_rate/action_model': 3.482180421399537e-05, 'learning_rate/action_query_proj': 3.482180421399537e-05, 'learning_rate/base': 8.705451053498842e-06, 'epoch': 0.53}) +[2026-06-30 03:32:03,505][starVLA.training.train_starvla][INFO] - Step 1234, Loss: {'train/loss': 0.0019631267059594393, 'train/grad_norm_pre_clip': 3.43727445602417, 'timing/data': 0.00019821408204734325, 'timing/model': 1.2292354661040008, 'learning_rate/action_model': 3.4751420991717256e-05, 'learning_rate/action_query_proj': 3.4751420991717256e-05, 'learning_rate/base': 8.687855247929314e-06, 'epoch': 0.53}) +[2026-06-30 03:32:36,181][starVLA.training.train_starvla][INFO] - Step 1235, Loss: {'train/loss': 0.0005748065887019038, 'train/grad_norm_pre_clip': 3.3216984272003174, 'timing/data': 0.00024962006136775017, 'timing/model': 1.2286740969866514, 'learning_rate/action_model': 3.468108032859319e-05, 'learning_rate/action_query_proj': 3.468108032859319e-05, 'learning_rate/base': 8.670270082148298e-06, 'epoch': 0.53}) +[2026-06-30 03:33:08,794][starVLA.training.train_starvla][INFO] - Step 1236, Loss: {'train/loss': 8.851220627548173e-06, 'train/grad_norm_pre_clip': 1.4733123779296875, 'timing/data': 0.0002702998463064432, 'timing/model': 1.22947983792983, 'learning_rate/action_model': 3.461078239818181e-05, 'learning_rate/action_query_proj': 3.461078239818181e-05, 'learning_rate/base': 8.652695599545452e-06, 'epoch': 0.53}) +[2026-06-30 03:33:41,482][starVLA.training.train_starvla][INFO] - Step 1237, Loss: {'train/loss': 1.6122630768222734e-05, 'train/grad_norm_pre_clip': 2.2530550956726074, 'timing/data': 0.00019085989333689213, 'timing/model': 1.2316237918566912, 'learning_rate/action_model': 3.454052737393624e-05, 'learning_rate/action_query_proj': 3.454052737393624e-05, 'learning_rate/base': 8.63513184348406e-06, 'epoch': 0.53}) +[2026-06-30 03:34:14,132][starVLA.training.train_starvla][INFO] - Step 1238, Loss: {'train/loss': 0.10776498913764954, 'train/grad_norm_pre_clip': 2.238262891769409, 'timing/data': 0.0003302029799669981, 'timing/model': 1.2285047001205385, 'learning_rate/action_model': 3.447031542920379e-05, 'learning_rate/action_query_proj': 3.447031542920379e-05, 'learning_rate/base': 8.617578857300947e-06, 'epoch': 0.53}) +[2026-06-30 03:34:46,808][starVLA.training.train_starvla][INFO] - Step 1239, Loss: {'train/loss': 4.771121894009411e-05, 'train/grad_norm_pre_clip': 1.9583345651626587, 'timing/data': 0.00021979492157697678, 'timing/model': 1.2356960051693022, 'learning_rate/action_model': 3.440014673722543e-05, 'learning_rate/action_query_proj': 3.440014673722543e-05, 'learning_rate/base': 8.600036684306357e-06, 'epoch': 0.53}) +[2026-06-30 03:35:19,440][starVLA.training.train_starvla][INFO] - Step 1240, Loss: {'train/loss': 0.3929819166660309, 'train/grad_norm_pre_clip': 3.9093806743621826, 'timing/data': 0.0002988839987665415, 'timing/model': 1.2312524928711355, 'learning_rate/action_model': 3.433002147113546e-05, 'learning_rate/action_query_proj': 3.433002147113546e-05, 'learning_rate/base': 8.582505367783866e-06, 'epoch': 0.53}) +[2026-06-30 03:35:52,090][starVLA.training.train_starvla][INFO] - Step 1241, Loss: {'train/loss': 0.18108010292053223, 'train/grad_norm_pre_clip': 3.1338257789611816, 'timing/data': 0.0002957081887871027, 'timing/model': 1.229980543954298, 'learning_rate/action_model': 3.425993980396097e-05, 'learning_rate/action_query_proj': 3.425993980396097e-05, 'learning_rate/base': 8.564984950990242e-06, 'epoch': 0.53}) +[2026-06-30 03:36:24,735][starVLA.training.train_starvla][INFO] - Step 1242, Loss: {'train/loss': 0.0030221743509173393, 'train/grad_norm_pre_clip': 2.1025443077087402, 'timing/data': 0.0001966310665011406, 'timing/model': 1.2253205438610166, 'learning_rate/action_model': 3.418990190862156e-05, 'learning_rate/action_query_proj': 3.418990190862156e-05, 'learning_rate/base': 8.54747547715539e-06, 'epoch': 0.53}) +[2026-06-30 03:36:57,396][starVLA.training.train_starvla][INFO] - Step 1243, Loss: {'train/loss': 0.001093194936402142, 'train/grad_norm_pre_clip': 4.50520133972168, 'timing/data': 0.00026552099734544754, 'timing/model': 1.2292923361528665, 'learning_rate/action_model': 3.411990795792871e-05, 'learning_rate/action_query_proj': 3.411990795792871e-05, 'learning_rate/base': 8.529976989482177e-06, 'epoch': 0.53}) +[2026-06-30 03:37:30,035][starVLA.training.train_starvla][INFO] - Step 1244, Loss: {'train/loss': 0.011688482947647572, 'train/grad_norm_pre_clip': 3.3013947010040283, 'timing/data': 0.0003023210447281599, 'timing/model': 1.227968449005857, 'learning_rate/action_model': 3.404995812458558e-05, 'learning_rate/action_query_proj': 3.404995812458558e-05, 'learning_rate/base': 8.512489531146396e-06, 'epoch': 0.54}) +[2026-06-30 03:38:02,644][starVLA.training.train_starvla][INFO] - Step 1245, Loss: {'train/loss': 0.0010378543520346284, 'train/grad_norm_pre_clip': 2.0201382637023926, 'timing/data': 0.00019885506480932236, 'timing/model': 1.2381107010878623, 'learning_rate/action_model': 3.3980052581186404e-05, 'learning_rate/action_query_proj': 3.3980052581186404e-05, 'learning_rate/base': 8.495013145296601e-06, 'epoch': 0.54}) +[2026-06-30 03:38:35,509][starVLA.training.train_starvla][INFO] - Step 1246, Loss: {'train/loss': 2.252972626592964e-05, 'train/grad_norm_pre_clip': 1.5260050296783447, 'timing/data': 0.00020498596131801605, 'timing/model': 1.2288260168861598, 'learning_rate/action_model': 3.3910191500216196e-05, 'learning_rate/action_query_proj': 3.3910191500216196e-05, 'learning_rate/base': 8.477547875054049e-06, 'epoch': 0.54}) +[2026-06-30 03:39:08,180][starVLA.training.train_starvla][INFO] - Step 1247, Loss: {'train/loss': 0.00038160078111104667, 'train/grad_norm_pre_clip': 2.8141980171203613, 'timing/data': 0.0001975430641323328, 'timing/model': 1.2352892979979515, 'learning_rate/action_model': 3.384037505405019e-05, 'learning_rate/action_query_proj': 3.384037505405019e-05, 'learning_rate/base': 8.460093763512547e-06, 'epoch': 0.54}) +[2026-06-30 03:39:40,869][starVLA.training.train_starvla][INFO] - Step 1248, Loss: {'train/loss': 0.15466123819351196, 'train/grad_norm_pre_clip': 2.3859498500823975, 'timing/data': 0.00027547916397452354, 'timing/model': 1.2316853110678494, 'learning_rate/action_model': 3.377060341495355e-05, 'learning_rate/action_query_proj': 3.377060341495355e-05, 'learning_rate/base': 8.442650853738388e-06, 'epoch': 0.54}) +[2026-06-30 03:40:13,439][starVLA.training.train_starvla][INFO] - Step 1249, Loss: {'train/loss': 5.262766717351042e-05, 'train/grad_norm_pre_clip': 2.6319637298583984, 'timing/data': 0.00025853817351162434, 'timing/model': 1.2303281109780073, 'learning_rate/action_model': 3.370087675508085e-05, 'learning_rate/action_query_proj': 3.370087675508085e-05, 'learning_rate/base': 8.425219188770212e-06, 'epoch': 0.54}) +[2026-06-30 03:40:46,100][starVLA.training.train_starvla][INFO] - Step 1250, Loss: {'train/loss': 0.00730097247287631, 'train/grad_norm_pre_clip': 2.6088249683380127, 'timing/data': 0.00021049706265330315, 'timing/model': 1.2282712101005018, 'learning_rate/action_model': 3.3631195246475695e-05, 'learning_rate/action_query_proj': 3.3631195246475695e-05, 'learning_rate/base': 8.407798811618924e-06, 'epoch': 0.54}) +[2026-06-30 03:41:18,747][starVLA.training.train_starvla][INFO] - Step 1251, Loss: {'train/loss': 0.013191154226660728, 'train/grad_norm_pre_clip': 3.035522937774658, 'timing/data': 0.00026837713085114956, 'timing/model': 1.231987060047686, 'learning_rate/action_model': 3.356155906107028e-05, 'learning_rate/action_query_proj': 3.356155906107028e-05, 'learning_rate/base': 8.39038976526757e-06, 'epoch': 0.54}) +[2026-06-30 03:41:51,442][starVLA.training.train_starvla][INFO] - Step 1252, Loss: {'train/loss': 0.03455360606312752, 'train/grad_norm_pre_clip': 1.955315351486206, 'timing/data': 0.00019800406880676746, 'timing/model': 1.2276761389803141, 'learning_rate/action_model': 3.3491968370684965e-05, 'learning_rate/action_query_proj': 3.3491968370684965e-05, 'learning_rate/base': 8.372992092671241e-06, 'epoch': 0.54}) +[2026-06-30 03:42:24,106][starVLA.training.train_starvla][INFO] - Step 1253, Loss: {'train/loss': 0.4390045702457428, 'train/grad_norm_pre_clip': 3.6280064582824707, 'timing/data': 0.00021881284192204475, 'timing/model': 1.2315173130482435, 'learning_rate/action_model': 3.342242334702788e-05, 'learning_rate/action_query_proj': 3.342242334702788e-05, 'learning_rate/base': 8.35560583675697e-06, 'epoch': 0.54}) +[2026-06-30 03:42:56,810][starVLA.training.train_starvla][INFO] - Step 1254, Loss: {'train/loss': 0.14181354641914368, 'train/grad_norm_pre_clip': 2.409742593765259, 'timing/data': 0.0003327978774905205, 'timing/model': 1.2267668738495559, 'learning_rate/action_model': 3.335292416169443e-05, 'learning_rate/action_query_proj': 3.335292416169443e-05, 'learning_rate/base': 8.338231040423608e-06, 'epoch': 0.54}) +[2026-06-30 03:43:29,522][starVLA.training.train_starvla][INFO] - Step 1255, Loss: {'train/loss': 1.0549841135798488e-05, 'train/grad_norm_pre_clip': 2.6369388103485107, 'timing/data': 0.00030665891245007515, 'timing/model': 1.2323267119936645, 'learning_rate/action_model': 3.328347098616698e-05, 'learning_rate/action_query_proj': 3.328347098616698e-05, 'learning_rate/base': 8.320867746541744e-06, 'epoch': 0.54}) +[2026-06-30 03:44:02,168][starVLA.training.train_starvla][INFO] - Step 1256, Loss: {'train/loss': 0.0013298513367772102, 'train/grad_norm_pre_clip': 2.113433361053467, 'timing/data': 0.0002137729898095131, 'timing/model': 1.2299474640749395, 'learning_rate/action_model': 3.321406399181428e-05, 'learning_rate/action_query_proj': 3.321406399181428e-05, 'learning_rate/base': 8.30351599795357e-06, 'epoch': 0.54}) +[2026-06-30 03:44:34,836][starVLA.training.train_starvla][INFO] - Step 1257, Loss: {'train/loss': 0.0039056080859154463, 'train/grad_norm_pre_clip': 1.6702489852905273, 'timing/data': 0.0002845372073352337, 'timing/model': 1.2345868449192494, 'learning_rate/action_model': 3.314470334989127e-05, 'learning_rate/action_query_proj': 3.314470334989127e-05, 'learning_rate/base': 8.286175837472817e-06, 'epoch': 0.54}) +[2026-06-30 03:45:07,483][starVLA.training.train_starvla][INFO] - Step 1258, Loss: {'train/loss': 0.0052116261795163155, 'train/grad_norm_pre_clip': 2.6226537227630615, 'timing/data': 0.00021211989223957062, 'timing/model': 1.2300180371385068, 'learning_rate/action_model': 3.307538923153835e-05, 'learning_rate/action_query_proj': 3.307538923153835e-05, 'learning_rate/base': 8.268847307884587e-06, 'epoch': 0.54}) +[2026-06-30 03:45:40,148][starVLA.training.train_starvla][INFO] - Step 1259, Loss: {'train/loss': 0.006033034063875675, 'train/grad_norm_pre_clip': 4.991316795349121, 'timing/data': 0.00019936682656407356, 'timing/model': 1.2281488708686084, 'learning_rate/action_model': 3.300612180778129e-05, 'learning_rate/action_query_proj': 3.300612180778129e-05, 'learning_rate/base': 8.251530451945323e-06, 'epoch': 0.54}) +[2026-06-30 03:46:12,820][starVLA.training.train_starvla][INFO] - Step 1260, Loss: {'train/loss': 0.001021582167595625, 'train/grad_norm_pre_clip': 1.953713297843933, 'timing/data': 0.00021414412185549736, 'timing/model': 1.2211672919802368, 'learning_rate/action_model': 3.2936901249530535e-05, 'learning_rate/action_query_proj': 3.2936901249530535e-05, 'learning_rate/base': 8.234225312382634e-06, 'epoch': 0.54}) +[2026-06-30 03:46:45,460][starVLA.training.train_starvla][INFO] - Step 1261, Loss: {'train/loss': 0.8500597476959229, 'train/grad_norm_pre_clip': 1.870866060256958, 'timing/data': 0.00026001990772783756, 'timing/model': 1.2266176140401512, 'learning_rate/action_model': 3.286772772758095e-05, 'learning_rate/action_query_proj': 3.286772772758095e-05, 'learning_rate/base': 8.216931931895238e-06, 'epoch': 0.54}) +[2026-06-30 03:47:18,081][starVLA.training.train_starvla][INFO] - Step 1262, Loss: {'train/loss': 0.20434467494487762, 'train/grad_norm_pre_clip': 4.1758952140808105, 'timing/data': 0.00020332308486104012, 'timing/model': 1.227798209991306, 'learning_rate/action_model': 3.2798601412611295e-05, 'learning_rate/action_query_proj': 3.2798601412611295e-05, 'learning_rate/base': 8.199650353152824e-06, 'epoch': 0.54}) +[2026-06-30 03:47:50,709][starVLA.training.train_starvla][INFO] - Step 1263, Loss: {'train/loss': 4.193036147626117e-05, 'train/grad_norm_pre_clip': 2.2575840950012207, 'timing/data': 0.0002556119579821825, 'timing/model': 1.2313532761763781, 'learning_rate/action_model': 3.272952247518392e-05, 'learning_rate/action_query_proj': 3.272952247518392e-05, 'learning_rate/base': 8.18238061879598e-06, 'epoch': 0.54}) +[2026-06-30 03:48:23,426][starVLA.training.train_starvla][INFO] - Step 1264, Loss: {'train/loss': 0.0015466449549421668, 'train/grad_norm_pre_clip': 2.5602641105651855, 'timing/data': 0.0002508629113435745, 'timing/model': 1.2298582519870251, 'learning_rate/action_model': 3.2660491085744195e-05, 'learning_rate/action_query_proj': 3.2660491085744195e-05, 'learning_rate/base': 8.165122771436049e-06, 'epoch': 0.54}) +[2026-06-30 03:48:56,070][starVLA.training.train_starvla][INFO] - Step 1265, Loss: {'train/loss': 0.009462404064834118, 'train/grad_norm_pre_clip': 0.9912509918212891, 'timing/data': 0.00020454591140151024, 'timing/model': 1.230363303096965, 'learning_rate/action_model': 3.259150741462024e-05, 'learning_rate/action_query_proj': 3.259150741462024e-05, 'learning_rate/base': 8.14787685365506e-06, 'epoch': 0.54}) +[2026-06-30 03:49:28,755][starVLA.training.train_starvla][INFO] - Step 1266, Loss: {'train/loss': 0.055608414113521576, 'train/grad_norm_pre_clip': 4.605697154998779, 'timing/data': 0.00046343496069312096, 'timing/model': 1.2311147258151323, 'learning_rate/action_model': 3.25225716320224e-05, 'learning_rate/action_query_proj': 3.25225716320224e-05, 'learning_rate/base': 8.1306429080056e-06, 'epoch': 0.54}) +[2026-06-30 03:50:01,421][starVLA.training.train_starvla][INFO] - Step 1267, Loss: {'train/loss': 0.009154172614216805, 'train/grad_norm_pre_clip': 1.8577948808670044, 'timing/data': 0.00021247100085020065, 'timing/model': 1.2311451949644834, 'learning_rate/action_model': 3.245368390804288e-05, 'learning_rate/action_query_proj': 3.245368390804288e-05, 'learning_rate/base': 8.11342097701072e-06, 'epoch': 0.55}) +[2026-06-30 03:50:34,094][starVLA.training.train_starvla][INFO] - Step 1268, Loss: {'train/loss': 5.364398020901717e-06, 'train/grad_norm_pre_clip': 2.2912728786468506, 'timing/data': 0.0002309461124241352, 'timing/model': 1.22916594799608, 'learning_rate/action_model': 3.238484441265525e-05, 'learning_rate/action_query_proj': 3.238484441265525e-05, 'learning_rate/base': 8.096211103163813e-06, 'epoch': 0.55}) +[2026-06-30 03:51:06,709][starVLA.training.train_starvla][INFO] - Step 1269, Loss: {'train/loss': 1.1205553164472803e-05, 'train/grad_norm_pre_clip': 1.7610526084899902, 'timing/data': 0.0002002681139856577, 'timing/model': 1.2285648200195283, 'learning_rate/action_model': 3.2316053315714155e-05, 'learning_rate/action_query_proj': 3.2316053315714155e-05, 'learning_rate/base': 8.079013328928539e-06, 'epoch': 0.55}) +[2026-06-30 03:51:39,370][starVLA.training.train_starvla][INFO] - Step 1270, Loss: {'train/loss': 0.0001971387246157974, 'train/grad_norm_pre_clip': 1.6700336933135986, 'timing/data': 0.00020326301455497742, 'timing/model': 1.233101475983858, 'learning_rate/action_model': 3.224731078695478e-05, 'learning_rate/action_query_proj': 3.224731078695478e-05, 'learning_rate/base': 8.061827696738694e-06, 'epoch': 0.55}) +[2026-06-30 03:52:12,007][starVLA.training.train_starvla][INFO] - Step 1271, Loss: {'train/loss': 0.041018567979335785, 'train/grad_norm_pre_clip': 2.1210873126983643, 'timing/data': 0.0002944059669971466, 'timing/model': 1.232640495058149, 'learning_rate/action_model': 3.217861699599248e-05, 'learning_rate/action_query_proj': 3.217861699599248e-05, 'learning_rate/base': 8.04465424899812e-06, 'epoch': 0.55}) +[2026-06-30 03:52:44,658][starVLA.training.train_starvla][INFO] - Step 1272, Loss: {'train/loss': 0.001541665056720376, 'train/grad_norm_pre_clip': 1.667276382446289, 'timing/data': 0.00019715283997356892, 'timing/model': 1.2324183799792081, 'learning_rate/action_model': 3.210997211232236e-05, 'learning_rate/action_query_proj': 3.210997211232236e-05, 'learning_rate/base': 8.02749302808059e-06, 'epoch': 0.55}) +[2026-06-30 03:53:17,303][starVLA.training.train_starvla][INFO] - Step 1273, Loss: {'train/loss': 0.004001552704721689, 'train/grad_norm_pre_clip': 1.1645350456237793, 'timing/data': 0.00019404594786465168, 'timing/model': 1.230886427918449, 'learning_rate/action_model': 3.2041376305318856e-05, 'learning_rate/action_query_proj': 3.2041376305318856e-05, 'learning_rate/base': 8.010344076329714e-06, 'epoch': 0.55}) +[2026-06-30 03:53:49,945][starVLA.training.train_starvla][INFO] - Step 1274, Loss: {'train/loss': 5.5011299991747364e-05, 'train/grad_norm_pre_clip': 2.8573925495147705, 'timing/data': 0.00019702105782926083, 'timing/model': 1.2327031721360981, 'learning_rate/action_model': 3.197282974423528e-05, 'learning_rate/action_query_proj': 3.197282974423528e-05, 'learning_rate/base': 7.99320743605882e-06, 'epoch': 0.55}) +[2026-06-30 03:54:22,616][starVLA.training.train_starvla][INFO] - Step 1275, Loss: {'train/loss': 0.00041748370858840644, 'train/grad_norm_pre_clip': 2.394890069961548, 'timing/data': 0.00025380891747772694, 'timing/model': 1.2267813740763813, 'learning_rate/action_model': 3.190433259820346e-05, 'learning_rate/action_query_proj': 3.190433259820346e-05, 'learning_rate/base': 7.976083149550865e-06, 'epoch': 0.55}) +[2026-06-30 03:54:55,264][starVLA.training.train_starvla][INFO] - Step 1276, Loss: {'train/loss': 0.00044346824870444834, 'train/grad_norm_pre_clip': 2.6119589805603027, 'timing/data': 0.0002928429748862982, 'timing/model': 1.231444027973339, 'learning_rate/action_model': 3.18358850362333e-05, 'learning_rate/action_query_proj': 3.18358850362333e-05, 'learning_rate/base': 7.958971259058325e-06, 'epoch': 0.55}) +[2026-06-30 03:55:27,926][starVLA.training.train_starvla][INFO] - Step 1277, Loss: {'train/loss': 0.001999236410483718, 'train/grad_norm_pre_clip': 2.9967503547668457, 'timing/data': 0.00019832397811114788, 'timing/model': 1.2287378991022706, 'learning_rate/action_model': 3.176748722721237e-05, 'learning_rate/action_query_proj': 3.176748722721237e-05, 'learning_rate/base': 7.941871806803092e-06, 'epoch': 0.55}) +[2026-06-30 03:56:00,579][starVLA.training.train_starvla][INFO] - Step 1278, Loss: {'train/loss': 0.03950818255543709, 'train/grad_norm_pre_clip': 0.9928359985351562, 'timing/data': 0.0002830440644174814, 'timing/model': 1.228885818971321, 'learning_rate/action_model': 3.169913933990545e-05, 'learning_rate/action_query_proj': 3.169913933990545e-05, 'learning_rate/base': 7.924784834976362e-06, 'epoch': 0.55}) +[2026-06-30 03:56:33,223][starVLA.training.train_starvla][INFO] - Step 1279, Loss: {'train/loss': 0.05271226540207863, 'train/grad_norm_pre_clip': 1.4962133169174194, 'timing/data': 0.00020754081197082996, 'timing/model': 1.2388469229917973, 'learning_rate/action_model': 3.163084154295416e-05, 'learning_rate/action_query_proj': 3.163084154295416e-05, 'learning_rate/base': 7.90771038573854e-06, 'epoch': 0.55}) +[2026-06-30 03:57:05,891][starVLA.training.train_starvla][INFO] - Step 1280, Loss: {'train/loss': 0.01010857243090868, 'train/grad_norm_pre_clip': 1.544400691986084, 'timing/data': 0.00020235218107700348, 'timing/model': 1.2313711140304804, 'learning_rate/action_model': 3.156259400487651e-05, 'learning_rate/action_query_proj': 3.156259400487651e-05, 'learning_rate/base': 7.890648501219127e-06, 'epoch': 0.55}) +[2026-06-30 03:57:38,500][starVLA.training.train_starvla][INFO] - Step 1281, Loss: {'train/loss': 0.12240730971097946, 'train/grad_norm_pre_clip': 2.8594584465026855, 'timing/data': 0.0003382780123502016, 'timing/model': 1.2302820710465312, 'learning_rate/action_model': 3.149439689406655e-05, 'learning_rate/action_query_proj': 3.149439689406655e-05, 'learning_rate/base': 7.873599223516637e-06, 'epoch': 0.55}) +[2026-06-30 03:58:11,159][starVLA.training.train_starvla][INFO] - Step 1282, Loss: {'train/loss': 0.00017167656915262341, 'train/grad_norm_pre_clip': 1.43441641330719, 'timing/data': 0.0004513419698923826, 'timing/model': 1.2321448200382292, 'learning_rate/action_model': 3.142625037879384e-05, 'learning_rate/action_query_proj': 3.142625037879384e-05, 'learning_rate/base': 7.85656259469846e-06, 'epoch': 0.55}) +[2026-06-30 03:58:43,813][starVLA.training.train_starvla][INFO] - Step 1283, Loss: {'train/loss': 0.012534002773463726, 'train/grad_norm_pre_clip': 1.5970722436904907, 'timing/data': 0.000275379978120327, 'timing/model': 1.2316066331695765, 'learning_rate/action_model': 3.135815462720316e-05, 'learning_rate/action_query_proj': 3.135815462720316e-05, 'learning_rate/base': 7.83953865680079e-06, 'epoch': 0.55}) +[2026-06-30 03:59:16,469][starVLA.training.train_starvla][INFO] - Step 1284, Loss: {'train/loss': 0.0396140031516552, 'train/grad_norm_pre_clip': 3.3302829265594482, 'timing/data': 0.00032103597186505795, 'timing/model': 1.2270504750777036, 'learning_rate/action_model': 3.129010980731396e-05, 'learning_rate/action_query_proj': 3.129010980731396e-05, 'learning_rate/base': 7.82252745182849e-06, 'epoch': 0.55}) +[2026-06-30 03:59:49,144][starVLA.training.train_starvla][INFO] - Step 1285, Loss: {'train/loss': 1.001345481199678e-05, 'train/grad_norm_pre_clip': 2.0324692726135254, 'timing/data': 0.00023857993073761463, 'timing/model': 1.2322552690748125, 'learning_rate/action_model': 3.122211608702015e-05, 'learning_rate/action_query_proj': 3.122211608702015e-05, 'learning_rate/base': 7.805529021755038e-06, 'epoch': 0.55}) +[2026-06-30 04:00:21,848][starVLA.training.train_starvla][INFO] - Step 1286, Loss: {'train/loss': 0.010241563431918621, 'train/grad_norm_pre_clip': 2.0524816513061523, 'timing/data': 0.0002960790880024433, 'timing/model': 1.2303228271193802, 'learning_rate/action_model': 3.115417363408938e-05, 'learning_rate/action_query_proj': 3.115417363408938e-05, 'learning_rate/base': 7.788543408522345e-06, 'epoch': 0.55}) +[2026-06-30 04:00:54,524][starVLA.training.train_starvla][INFO] - Step 1287, Loss: {'train/loss': 0.20724846422672272, 'train/grad_norm_pre_clip': 2.2989182472229004, 'timing/data': 0.00031506503000855446, 'timing/model': 1.2293250318616629, 'learning_rate/action_model': 3.108628261616297e-05, 'learning_rate/action_query_proj': 3.108628261616297e-05, 'learning_rate/base': 7.771570654040742e-06, 'epoch': 0.55}) +[2026-06-30 04:01:27,198][starVLA.training.train_starvla][INFO] - Step 1288, Loss: {'train/loss': 0.01695498637855053, 'train/grad_norm_pre_clip': 2.1484670639038086, 'timing/data': 0.00026857713237404823, 'timing/model': 1.217237361939624, 'learning_rate/action_model': 3.101844320075523e-05, 'learning_rate/action_query_proj': 3.101844320075523e-05, 'learning_rate/base': 7.754610800188808e-06, 'epoch': 0.55}) +[2026-06-30 04:01:59,867][starVLA.training.train_starvla][INFO] - Step 1289, Loss: {'train/loss': 0.08306211978197098, 'train/grad_norm_pre_clip': 3.1415412425994873, 'timing/data': 0.00020014680922031403, 'timing/model': 1.2277078600600362, 'learning_rate/action_model': 3.0950655555253175e-05, 'learning_rate/action_query_proj': 3.0950655555253175e-05, 'learning_rate/base': 7.737663888813294e-06, 'epoch': 0.55}) +[2026-06-30 04:02:32,504][starVLA.training.train_starvla][INFO] - Step 1290, Loss: {'train/loss': 0.0043848599307239056, 'train/grad_norm_pre_clip': 1.7755175828933716, 'timing/data': 0.00023966189473867416, 'timing/model': 1.231605056906119, 'learning_rate/action_model': 3.088291984691607e-05, 'learning_rate/action_query_proj': 3.088291984691607e-05, 'learning_rate/base': 7.720729961729018e-06, 'epoch': 0.56}) +[2026-06-30 04:03:05,154][starVLA.training.train_starvla][INFO] - Step 1291, Loss: {'train/loss': 2.7655163648887537e-05, 'train/grad_norm_pre_clip': 1.9216406345367432, 'timing/data': 0.000206309137865901, 'timing/model': 1.236165767069906, 'learning_rate/action_model': 3.081523624287507e-05, 'learning_rate/action_query_proj': 3.081523624287507e-05, 'learning_rate/base': 7.703809060718768e-06, 'epoch': 0.56}) +[2026-06-30 04:03:37,791][starVLA.training.train_starvla][INFO] - Step 1292, Loss: {'train/loss': 0.011621523648500443, 'train/grad_norm_pre_clip': 1.7868859767913818, 'timing/data': 0.0002581870649009943, 'timing/model': 1.227608506102115, 'learning_rate/action_model': 3.0747604910132736e-05, 'learning_rate/action_query_proj': 3.0747604910132736e-05, 'learning_rate/base': 7.686901227533184e-06, 'epoch': 0.56}) +[2026-06-30 04:04:10,411][starVLA.training.train_starvla][INFO] - Step 1293, Loss: {'train/loss': 0.19071932137012482, 'train/grad_norm_pre_clip': 3.129631757736206, 'timing/data': 0.00024030311033129692, 'timing/model': 1.2320998988579959, 'learning_rate/action_model': 3.068002601556263e-05, 'learning_rate/action_query_proj': 3.068002601556263e-05, 'learning_rate/base': 7.670006503890658e-06, 'epoch': 0.56}) +[2026-06-30 04:04:43,086][starVLA.training.train_starvla][INFO] - Step 1294, Loss: {'train/loss': 0.00020105672592762858, 'train/grad_norm_pre_clip': 4.280309200286865, 'timing/data': 0.00019542896188795567, 'timing/model': 1.230208900058642, 'learning_rate/action_model': 3.061249972590898e-05, 'learning_rate/action_query_proj': 3.061249972590898e-05, 'learning_rate/base': 7.653124931477245e-06, 'epoch': 0.56}) +[2026-06-30 04:05:15,712][starVLA.training.train_starvla][INFO] - Step 1295, Loss: {'train/loss': 1.6093238173198188e-06, 'train/grad_norm_pre_clip': 1.527849793434143, 'timing/data': 0.00020103901624679565, 'timing/model': 1.228400006890297, 'learning_rate/action_model': 3.0545026207786185e-05, 'learning_rate/action_query_proj': 3.0545026207786185e-05, 'learning_rate/base': 7.636256551946546e-06, 'epoch': 0.56}) +[2026-06-30 04:05:48,405][starVLA.training.train_starvla][INFO] - Step 1296, Loss: {'train/loss': 3.0100259209575597e-06, 'train/grad_norm_pre_clip': 2.2653989791870117, 'timing/data': 0.0004601581022143364, 'timing/model': 1.2280676618684083, 'learning_rate/action_model': 3.0477605627678456e-05, 'learning_rate/action_query_proj': 3.0477605627678456e-05, 'learning_rate/base': 7.619401406919614e-06, 'epoch': 0.56}) +[2026-06-30 04:06:21,059][starVLA.training.train_starvla][INFO] - Step 1297, Loss: {'train/loss': 0.006574879866093397, 'train/grad_norm_pre_clip': 1.7144880294799805, 'timing/data': 0.0003466841299086809, 'timing/model': 1.2304624929092824, 'learning_rate/action_model': 3.0410238151939364e-05, 'learning_rate/action_query_proj': 3.0410238151939364e-05, 'learning_rate/base': 7.602559537984841e-06, 'epoch': 0.56}) +[2026-06-30 04:06:53,720][starVLA.training.train_starvla][INFO] - Step 1298, Loss: {'train/loss': 0.11570713669061661, 'train/grad_norm_pre_clip': 4.458329677581787, 'timing/data': 0.0003022509627044201, 'timing/model': 1.2302485429681838, 'learning_rate/action_model': 3.034292394679143e-05, 'learning_rate/action_query_proj': 3.034292394679143e-05, 'learning_rate/base': 7.585730986697857e-06, 'epoch': 0.56}) +[2026-06-30 04:07:26,565][starVLA.training.train_starvla][INFO] - Step 1299, Loss: {'train/loss': 0.0033761190716177225, 'train/grad_norm_pre_clip': 2.589900493621826, 'timing/data': 0.00026516104117035866, 'timing/model': 1.2320631109178066, 'learning_rate/action_model': 3.027566317832582e-05, 'learning_rate/action_query_proj': 3.027566317832582e-05, 'learning_rate/base': 7.568915794581455e-06, 'epoch': 0.56}) +[2026-06-30 04:07:59,243][starVLA.training.train_starvla][INFO] - Step 1300, Loss: {'train/loss': 0.09198158979415894, 'train/grad_norm_pre_clip': 3.423717975616455, 'timing/data': 0.00035222386941313744, 'timing/model': 1.2327322189230472, 'learning_rate/action_model': 3.0208456012501756e-05, 'learning_rate/action_query_proj': 3.0208456012501756e-05, 'learning_rate/base': 7.552114003125439e-06, 'epoch': 0.56}) +[2026-06-30 04:08:31,899][starVLA.training.train_starvla][INFO] - Step 1301, Loss: {'train/loss': 2.8014137569698505e-06, 'train/grad_norm_pre_clip': 4.897959232330322, 'timing/data': 0.0003657001070678234, 'timing/model': 1.2365739040542394, 'learning_rate/action_model': 3.014130261514625e-05, 'learning_rate/action_query_proj': 3.014130261514625e-05, 'learning_rate/base': 7.535325653786563e-06, 'epoch': 0.56}) +[2026-06-30 04:09:04,566][starVLA.training.train_starvla][INFO] - Step 1302, Loss: {'train/loss': 0.018365759402513504, 'train/grad_norm_pre_clip': 5.508665561676025, 'timing/data': 0.0001954887993633747, 'timing/model': 1.227988292928785, 'learning_rate/action_model': 3.007420315195361e-05, 'learning_rate/action_query_proj': 3.007420315195361e-05, 'learning_rate/base': 7.5185507879884025e-06, 'epoch': 0.56}) +[2026-06-30 04:09:37,203][starVLA.training.train_starvla][INFO] - Step 1303, Loss: {'train/loss': 0.1748490035533905, 'train/grad_norm_pre_clip': 2.319328784942627, 'timing/data': 0.0002567840274423361, 'timing/model': 1.2277532080188394, 'learning_rate/action_model': 3.0007157788485158e-05, 'learning_rate/action_query_proj': 3.0007157788485158e-05, 'learning_rate/base': 7.5017894471212895e-06, 'epoch': 0.56}) +[2026-06-30 04:10:09,871][starVLA.training.train_starvla][INFO] - Step 1304, Loss: {'train/loss': 0.28254634141921997, 'train/grad_norm_pre_clip': 1.33819580078125, 'timing/data': 0.0002809201832860708, 'timing/model': 1.2276637579780072, 'learning_rate/action_model': 2.9940166690168587e-05, 'learning_rate/action_query_proj': 2.9940166690168587e-05, 'learning_rate/base': 7.485041672542147e-06, 'epoch': 0.56}) +[2026-06-30 04:10:42,527][starVLA.training.train_starvla][INFO] - Step 1305, Loss: {'train/loss': 2.086160066028242e-06, 'train/grad_norm_pre_clip': 1.1567306518554688, 'timing/data': 0.00023857993073761463, 'timing/model': 1.2307095839641988, 'learning_rate/action_model': 2.9873230022297816e-05, 'learning_rate/action_query_proj': 2.9873230022297816e-05, 'learning_rate/base': 7.468307505574454e-06, 'epoch': 0.56}) +[2026-06-30 04:11:15,169][starVLA.training.train_starvla][INFO] - Step 1306, Loss: {'train/loss': 0.15830445289611816, 'train/grad_norm_pre_clip': 3.6444008350372314, 'timing/data': 0.0002083128783851862, 'timing/model': 1.2337561121676117, 'learning_rate/action_model': 2.9806347950032402e-05, 'learning_rate/action_query_proj': 2.9806347950032402e-05, 'learning_rate/base': 7.451586987508101e-06, 'epoch': 0.56}) +[2026-06-30 04:11:47,810][starVLA.training.train_starvla][INFO] - Step 1307, Loss: {'train/loss': 0.02515415847301483, 'train/grad_norm_pre_clip': 3.2589972019195557, 'timing/data': 0.00020374497398734093, 'timing/model': 1.2244642318692058, 'learning_rate/action_model': 2.9739520638397216e-05, 'learning_rate/action_query_proj': 2.9739520638397216e-05, 'learning_rate/base': 7.434880159599304e-06, 'epoch': 0.56}) +[2026-06-30 04:12:20,425][starVLA.training.train_starvla][INFO] - Step 1308, Loss: {'train/loss': 0.3384707570075989, 'train/grad_norm_pre_clip': 2.4361448287963867, 'timing/data': 0.0002905079163610935, 'timing/model': 1.2307721350807697, 'learning_rate/action_model': 2.9672748252281997e-05, 'learning_rate/action_query_proj': 2.9672748252281997e-05, 'learning_rate/base': 7.418187063070499e-06, 'epoch': 0.56}) +[2026-06-30 04:12:53,085][starVLA.training.train_starvla][INFO] - Step 1309, Loss: {'train/loss': 7.75352309574373e-05, 'train/grad_norm_pre_clip': 1.2645026445388794, 'timing/data': 0.0002027119044214487, 'timing/model': 1.2289168059360236, 'learning_rate/action_model': 2.9606030956440994e-05, 'learning_rate/action_query_proj': 2.9606030956440994e-05, 'learning_rate/base': 7.4015077391102485e-06, 'epoch': 0.56}) +[2026-06-30 04:13:25,782][starVLA.training.train_starvla][INFO] - Step 1310, Loss: {'train/loss': 0.0835656076669693, 'train/grad_norm_pre_clip': 2.347151517868042, 'timing/data': 0.00019732187502086163, 'timing/model': 1.235419605858624, 'learning_rate/action_model': 2.9539368915492484e-05, 'learning_rate/action_query_proj': 2.9539368915492484e-05, 'learning_rate/base': 7.384842228873121e-06, 'epoch': 0.56}) +[2026-06-30 04:13:58,460][starVLA.training.train_starvla][INFO] - Step 1311, Loss: {'train/loss': 1.4672365188598633, 'train/grad_norm_pre_clip': 2.6890106201171875, 'timing/data': 0.00019400590099394321, 'timing/model': 1.233290382893756, 'learning_rate/action_model': 2.9472762293918426e-05, 'learning_rate/action_query_proj': 2.9472762293918426e-05, 'learning_rate/base': 7.368190573479607e-06, 'epoch': 0.56}) +[2026-06-30 04:14:31,041][starVLA.training.train_starvla][INFO] - Step 1312, Loss: {'train/loss': 0.024958578869700432, 'train/grad_norm_pre_clip': 1.9933900833129883, 'timing/data': 0.00025474000722169876, 'timing/model': 1.230065671959892, 'learning_rate/action_model': 2.9406211256064036e-05, 'learning_rate/action_query_proj': 2.9406211256064036e-05, 'learning_rate/base': 7.351552814016009e-06, 'epoch': 0.56}) +[2026-06-30 04:15:03,685][starVLA.training.train_starvla][INFO] - Step 1313, Loss: {'train/loss': 0.08838127553462982, 'train/grad_norm_pre_clip': 2.1110000610351562, 'timing/data': 0.0003185609821230173, 'timing/model': 1.2337290069554, 'learning_rate/action_model': 2.9339715966137398e-05, 'learning_rate/action_query_proj': 2.9339715966137398e-05, 'learning_rate/base': 7.3349289915343496e-06, 'epoch': 0.56}) +[2026-06-30 04:15:36,364][starVLA.training.train_starvla][INFO] - Step 1314, Loss: {'train/loss': 0.3882773816585541, 'train/grad_norm_pre_clip': 3.654306650161743, 'timing/data': 0.0002543199807405472, 'timing/model': 1.2311629399191588, 'learning_rate/action_model': 2.9273276588209014e-05, 'learning_rate/action_query_proj': 2.9273276588209014e-05, 'learning_rate/base': 7.3183191470522534e-06, 'epoch': 0.57}) +[2026-06-30 04:16:09,023][starVLA.training.train_starvla][INFO] - Step 1315, Loss: {'train/loss': 0.016338404268026352, 'train/grad_norm_pre_clip': 3.686457395553589, 'timing/data': 0.00019361614249646664, 'timing/model': 1.2323531499132514, 'learning_rate/action_model': 2.9206893286211468e-05, 'learning_rate/action_query_proj': 2.9206893286211468e-05, 'learning_rate/base': 7.301723321552867e-06, 'epoch': 0.57}) +[2026-06-30 04:16:41,700][starVLA.training.train_starvla][INFO] - Step 1316, Loss: {'train/loss': 0.0003197719343006611, 'train/grad_norm_pre_clip': 1.943996787071228, 'timing/data': 0.00020786188542842865, 'timing/model': 1.227555994875729, 'learning_rate/action_model': 2.9140566223938913e-05, 'learning_rate/action_query_proj': 2.9140566223938913e-05, 'learning_rate/base': 7.285141555984728e-06, 'epoch': 0.57}) +[2026-06-30 04:17:14,365][starVLA.training.train_starvla][INFO] - Step 1317, Loss: {'train/loss': 0.005019530188292265, 'train/grad_norm_pre_clip': 3.364016056060791, 'timing/data': 0.000296969898045063, 'timing/model': 1.2314288860652596, 'learning_rate/action_model': 2.907429556504684e-05, 'learning_rate/action_query_proj': 2.907429556504684e-05, 'learning_rate/base': 7.26857389126171e-06, 'epoch': 0.57}) +[2026-06-30 04:17:47,031][starVLA.training.train_starvla][INFO] - Step 1318, Loss: {'train/loss': 0.003564161015674472, 'train/grad_norm_pre_clip': 2.1533942222595215, 'timing/data': 0.0003292621113359928, 'timing/model': 1.2375083910301328, 'learning_rate/action_model': 2.9008081473051474e-05, 'learning_rate/action_query_proj': 2.9008081473051474e-05, 'learning_rate/base': 7.252020368262869e-06, 'epoch': 0.57}) +[2026-06-30 04:18:19,708][starVLA.training.train_starvla][INFO] - Step 1319, Loss: {'train/loss': 0.41575703024864197, 'train/grad_norm_pre_clip': 3.0257911682128906, 'timing/data': 0.0002481879200786352, 'timing/model': 1.2338544130325317, 'learning_rate/action_model': 2.8941924111329515e-05, 'learning_rate/action_query_proj': 2.8941924111329515e-05, 'learning_rate/base': 7.235481027832379e-06, 'epoch': 0.57}) +[2026-06-30 04:18:52,368][starVLA.training.train_starvla][INFO] - Step 1320, Loss: {'train/loss': 0.1480136513710022, 'train/grad_norm_pre_clip': 3.311735153198242, 'timing/data': 0.0002087140455842018, 'timing/model': 1.2268864940851927, 'learning_rate/action_model': 2.887582364311766e-05, 'learning_rate/action_query_proj': 2.887582364311766e-05, 'learning_rate/base': 7.218955910779415e-06, 'epoch': 0.57}) +[2026-06-30 04:19:25,038][starVLA.training.train_starvla][INFO] - Step 1321, Loss: {'train/loss': 7.152505986596225e-06, 'train/grad_norm_pre_clip': 3.3056178092956543, 'timing/data': 0.00020244205370545387, 'timing/model': 1.2321018690709025, 'learning_rate/action_model': 2.880978023151229e-05, 'learning_rate/action_query_proj': 2.880978023151229e-05, 'learning_rate/base': 7.202445057878073e-06, 'epoch': 0.57}) +[2026-06-30 04:19:57,705][starVLA.training.train_starvla][INFO] - Step 1322, Loss: {'train/loss': 0.0001352386170765385, 'train/grad_norm_pre_clip': 3.2775051593780518, 'timing/data': 0.000322938896715641, 'timing/model': 1.2252985909581184, 'learning_rate/action_model': 2.8743794039468897e-05, 'learning_rate/action_query_proj': 2.8743794039468897e-05, 'learning_rate/base': 7.185948509867224e-06, 'epoch': 0.57}) +[2026-06-30 04:20:30,336][starVLA.training.train_starvla][INFO] - Step 1323, Loss: {'train/loss': 0.49913981556892395, 'train/grad_norm_pre_clip': 4.407509803771973, 'timing/data': 0.0002222789917141199, 'timing/model': 1.2294963670428842, 'learning_rate/action_model': 2.8677865229801898e-05, 'learning_rate/action_query_proj': 2.8677865229801898e-05, 'learning_rate/base': 7.1694663074504746e-06, 'epoch': 0.57}) +[2026-06-30 04:21:02,994][starVLA.training.train_starvla][INFO] - Step 1324, Loss: {'train/loss': 0.0030179074965417385, 'train/grad_norm_pre_clip': 3.0326380729675293, 'timing/data': 0.00030676904134452343, 'timing/model': 1.2363833670970052, 'learning_rate/action_model': 2.8611993965184036e-05, 'learning_rate/action_query_proj': 2.8611993965184036e-05, 'learning_rate/base': 7.152998491296009e-06, 'epoch': 0.57}) +[2026-06-30 04:21:35,665][starVLA.training.train_starvla][INFO] - Step 1325, Loss: {'train/loss': 0.019628534093499184, 'train/grad_norm_pre_clip': 2.153750419616699, 'timing/data': 0.0004039318300783634, 'timing/model': 1.2288068910129368, 'learning_rate/action_model': 2.854618040814618e-05, 'learning_rate/action_query_proj': 2.854618040814618e-05, 'learning_rate/base': 7.136545102036545e-06, 'epoch': 0.57}) +[2026-06-30 04:22:08,340][starVLA.training.train_starvla][INFO] - Step 1326, Loss: {'train/loss': 8.4208048065193e-05, 'train/grad_norm_pre_clip': 3.297969341278076, 'timing/data': 0.00019506807439029217, 'timing/model': 1.2354497900232673, 'learning_rate/action_model': 2.8480424721076642e-05, 'learning_rate/action_query_proj': 2.8480424721076642e-05, 'learning_rate/base': 7.120106180269161e-06, 'epoch': 0.57}) +[2026-06-30 04:22:40,983][starVLA.training.train_starvla][INFO] - Step 1327, Loss: {'train/loss': 0.20541858673095703, 'train/grad_norm_pre_clip': 2.916004180908203, 'timing/data': 0.00020279292948544025, 'timing/model': 1.2306491809431463, 'learning_rate/action_model': 2.8414727066221115e-05, 'learning_rate/action_query_proj': 2.8414727066221115e-05, 'learning_rate/base': 7.103681766555279e-06, 'epoch': 0.57}) +[2026-06-30 04:23:13,661][starVLA.training.train_starvla][INFO] - Step 1328, Loss: {'train/loss': 0.05174838751554489, 'train/grad_norm_pre_clip': 3.5763299465179443, 'timing/data': 0.00020590890198946, 'timing/model': 1.2327569650951773, 'learning_rate/action_model': 2.8349087605682012e-05, 'learning_rate/action_query_proj': 2.8349087605682012e-05, 'learning_rate/base': 7.087271901420503e-06, 'epoch': 0.57}) +[2026-06-30 04:23:46,346][starVLA.training.train_starvla][INFO] - Step 1329, Loss: {'train/loss': 0.045483656227588654, 'train/grad_norm_pre_clip': 3.4183220863342285, 'timing/data': 0.0003383078146725893, 'timing/model': 1.2341823508031666, 'learning_rate/action_model': 2.8283506501418173e-05, 'learning_rate/action_query_proj': 2.8283506501418173e-05, 'learning_rate/base': 7.070876625354543e-06, 'epoch': 0.57}) +[2026-06-30 04:24:19,018][starVLA.training.train_starvla][INFO] - Step 1330, Loss: {'train/loss': 0.04332723841071129, 'train/grad_norm_pre_clip': 2.8337628841400146, 'timing/data': 0.0001975828781723976, 'timing/model': 1.2319103630725294, 'learning_rate/action_model': 2.8217983915244433e-05, 'learning_rate/action_query_proj': 2.8217983915244433e-05, 'learning_rate/base': 7.054495978811108e-06, 'epoch': 0.57}) +[2026-06-30 04:24:51,659][starVLA.training.train_starvla][INFO] - Step 1331, Loss: {'train/loss': 0.0023591448552906513, 'train/grad_norm_pre_clip': 1.5875319242477417, 'timing/data': 0.0004497179761528969, 'timing/model': 1.2317692451179028, 'learning_rate/action_model': 2.8152520008831308e-05, 'learning_rate/action_query_proj': 2.8152520008831308e-05, 'learning_rate/base': 7.038130002207827e-06, 'epoch': 0.57}) +[2026-06-30 04:25:24,344][starVLA.training.train_starvla][INFO] - Step 1332, Loss: {'train/loss': 1.6778536519268528e-05, 'train/grad_norm_pre_clip': 1.9255144596099854, 'timing/data': 0.00020081805996596813, 'timing/model': 1.2313757499214262, 'learning_rate/action_model': 2.808711494370444e-05, 'learning_rate/action_query_proj': 2.808711494370444e-05, 'learning_rate/base': 7.02177873592611e-06, 'epoch': 0.57}) +[2026-06-30 04:25:57,004][starVLA.training.train_starvla][INFO] - Step 1333, Loss: {'train/loss': 0.07581067830324173, 'train/grad_norm_pre_clip': 2.425952434539795, 'timing/data': 0.0002739769406616688, 'timing/model': 1.2290701898746192, 'learning_rate/action_model': 2.8021768881244353e-05, 'learning_rate/action_query_proj': 2.8021768881244353e-05, 'learning_rate/base': 7.005442220311088e-06, 'epoch': 0.57}) +[2026-06-30 04:26:29,647][starVLA.training.train_starvla][INFO] - Step 1334, Loss: {'train/loss': 0.12689109146595, 'train/grad_norm_pre_clip': 1.6566643714904785, 'timing/data': 0.00021068821661174297, 'timing/model': 1.233118565985933, 'learning_rate/action_model': 2.7956481982685918e-05, 'learning_rate/action_query_proj': 2.7956481982685918e-05, 'learning_rate/base': 6.9891204956714795e-06, 'epoch': 0.57}) +[2026-06-30 04:27:02,315][starVLA.training.train_starvla][INFO] - Step 1335, Loss: {'train/loss': 0.08939788490533829, 'train/grad_norm_pre_clip': 1.2044788599014282, 'timing/data': 0.00020590797066688538, 'timing/model': 1.2292590930592269, 'learning_rate/action_model': 2.7891254409118116e-05, 'learning_rate/action_query_proj': 2.7891254409118116e-05, 'learning_rate/base': 6.972813602279529e-06, 'epoch': 0.57}) +[2026-06-30 04:27:34,975][starVLA.training.train_starvla][INFO] - Step 1336, Loss: {'train/loss': 0.000317889207508415, 'train/grad_norm_pre_clip': 1.7128925323486328, 'timing/data': 0.00020587793551385403, 'timing/model': 1.2304778899997473, 'learning_rate/action_model': 2.782608632148348e-05, 'learning_rate/action_query_proj': 2.782608632148348e-05, 'learning_rate/base': 6.95652158037087e-06, 'epoch': 0.57}) +[2026-06-30 04:28:07,649][starVLA.training.train_starvla][INFO] - Step 1337, Loss: {'train/loss': 0.19829913973808289, 'train/grad_norm_pre_clip': 1.8878211975097656, 'timing/data': 0.00029526720754802227, 'timing/model': 1.22539894701913, 'learning_rate/action_model': 2.7760977880577794e-05, 'learning_rate/action_query_proj': 2.7760977880577794e-05, 'learning_rate/base': 6.9402444701444486e-06, 'epoch': 0.58}) +[2026-06-30 04:28:40,311][starVLA.training.train_starvla][INFO] - Step 1338, Loss: {'train/loss': 0.028755174949765205, 'train/grad_norm_pre_clip': 1.1510928869247437, 'timing/data': 0.00020713103003799915, 'timing/model': 1.2338512521237135, 'learning_rate/action_model': 2.769592924704963e-05, 'learning_rate/action_query_proj': 2.769592924704963e-05, 'learning_rate/base': 6.923982311762408e-06, 'epoch': 0.58}) +[2026-06-30 04:29:12,948][starVLA.training.train_starvla][INFO] - Step 1339, Loss: {'train/loss': 4.619348601409001e-06, 'train/grad_norm_pre_clip': 1.1278105974197388, 'timing/data': 0.00019391486421227455, 'timing/model': 1.2294353998731822, 'learning_rate/action_model': 2.7630940581400094e-05, 'learning_rate/action_query_proj': 2.7630940581400094e-05, 'learning_rate/base': 6.9077351453500235e-06, 'epoch': 0.58}) +[2026-06-30 04:29:45,630][starVLA.training.train_starvla][INFO] - Step 1340, Loss: {'train/loss': 0.00500616617500782, 'train/grad_norm_pre_clip': 3.2943081855773926, 'timing/data': 0.0002145951148122549, 'timing/model': 1.2312391120940447, 'learning_rate/action_model': 2.7566012043982175e-05, 'learning_rate/action_query_proj': 2.7566012043982175e-05, 'learning_rate/base': 6.891503010995544e-06, 'epoch': 0.58}) +[2026-06-30 04:30:18,291][starVLA.training.train_starvla][INFO] - Step 1341, Loss: {'train/loss': 0.004479552153497934, 'train/grad_norm_pre_clip': 3.307093381881714, 'timing/data': 0.00020647002384066582, 'timing/model': 1.2324133759830147, 'learning_rate/action_model': 2.750114379500065e-05, 'learning_rate/action_query_proj': 2.750114379500065e-05, 'learning_rate/base': 6.875285948750162e-06, 'epoch': 0.58}) +[2026-06-30 04:30:50,936][starVLA.training.train_starvla][INFO] - Step 1342, Loss: {'train/loss': 1.4489195346832275, 'train/grad_norm_pre_clip': 3.7044055461883545, 'timing/data': 0.00019573979079723358, 'timing/model': 1.2351891919970512, 'learning_rate/action_model': 2.7436335994511432e-05, 'learning_rate/action_query_proj': 2.7436335994511432e-05, 'learning_rate/base': 6.859083998627858e-06, 'epoch': 0.58}) +[2026-06-30 04:31:23,581][starVLA.training.train_starvla][INFO] - Step 1343, Loss: {'train/loss': 0.011502132751047611, 'train/grad_norm_pre_clip': 4.508075714111328, 'timing/data': 0.00025463011115789413, 'timing/model': 1.2283175978809595, 'learning_rate/action_model': 2.7371588802421383e-05, 'learning_rate/action_query_proj': 2.7371588802421383e-05, 'learning_rate/base': 6.842897200605346e-06, 'epoch': 0.58}) +[2026-06-30 04:31:56,201][starVLA.training.train_starvla][INFO] - Step 1344, Loss: {'train/loss': 0.007969524711370468, 'train/grad_norm_pre_clip': 2.430184841156006, 'timing/data': 0.00020066904835402966, 'timing/model': 1.2290079090744257, 'learning_rate/action_model': 2.7306902378487687e-05, 'learning_rate/action_query_proj': 2.7306902378487687e-05, 'learning_rate/base': 6.826725594621922e-06, 'epoch': 0.58}) +[2026-06-30 04:32:28,879][starVLA.training.train_starvla][INFO] - Step 1345, Loss: {'train/loss': 0.00017115470836870372, 'train/grad_norm_pre_clip': 2.145148992538452, 'timing/data': 0.00020999601110816002, 'timing/model': 1.2311781118623912, 'learning_rate/action_model': 2.7242276882317724e-05, 'learning_rate/action_query_proj': 2.7242276882317724e-05, 'learning_rate/base': 6.810569220579431e-06, 'epoch': 0.58}) +[2026-06-30 04:33:01,562][starVLA.training.train_starvla][INFO] - Step 1346, Loss: {'train/loss': 0.010648385621607304, 'train/grad_norm_pre_clip': 1.8959318399429321, 'timing/data': 0.0001970019657164812, 'timing/model': 1.2347393590025604, 'learning_rate/action_model': 2.717771247336846e-05, 'learning_rate/action_query_proj': 2.717771247336846e-05, 'learning_rate/base': 6.794428118342115e-06, 'epoch': 0.58}) +[2026-06-30 04:33:34,231][starVLA.training.train_starvla][INFO] - Step 1347, Loss: {'train/loss': 0.0036678239703178406, 'train/grad_norm_pre_clip': 2.390475034713745, 'timing/data': 0.00026068207807838917, 'timing/model': 1.2297881678678095, 'learning_rate/action_model': 2.711320931094616e-05, 'learning_rate/action_query_proj': 2.711320931094616e-05, 'learning_rate/base': 6.77830232773654e-06, 'epoch': 0.58}) +[2026-06-30 04:34:06,883][starVLA.training.train_starvla][INFO] - Step 1348, Loss: {'train/loss': 0.21829073131084442, 'train/grad_norm_pre_clip': 2.075253963470459, 'timing/data': 0.00030216993764042854, 'timing/model': 1.228203017031774, 'learning_rate/action_model': 2.704876755420594e-05, 'learning_rate/action_query_proj': 2.704876755420594e-05, 'learning_rate/base': 6.762191888551485e-06, 'epoch': 0.58}) +[2026-06-30 04:34:39,534][starVLA.training.train_starvla][INFO] - Step 1349, Loss: {'train/loss': 0.4629666805267334, 'train/grad_norm_pre_clip': 2.6741673946380615, 'timing/data': 0.00025573186576366425, 'timing/model': 1.227100407006219, 'learning_rate/action_model': 2.698438736215147e-05, 'learning_rate/action_query_proj': 2.698438736215147e-05, 'learning_rate/base': 6.746096840537868e-06, 'epoch': 0.58}) +[2026-06-30 04:35:12,180][starVLA.training.train_starvla][INFO] - Step 1350, Loss: {'train/loss': 0.19910001754760742, 'train/grad_norm_pre_clip': 1.2103534936904907, 'timing/data': 0.0003000160213559866, 'timing/model': 1.2297199280001223, 'learning_rate/action_model': 2.692006889363446e-05, 'learning_rate/action_query_proj': 2.692006889363446e-05, 'learning_rate/base': 6.730017223408615e-06, 'epoch': 0.58}) +[2026-06-30 04:35:45,021][starVLA.training.train_starvla][INFO] - Step 1351, Loss: {'train/loss': 2.4526372726541013e-05, 'train/grad_norm_pre_clip': 2.068653106689453, 'timing/data': 0.0002880529500544071, 'timing/model': 1.2303773979656398, 'learning_rate/action_model': 2.685581230735433e-05, 'learning_rate/action_query_proj': 2.685581230735433e-05, 'learning_rate/base': 6.713953076838582e-06, 'epoch': 0.58}) +[2026-06-30 04:36:17,681][starVLA.training.train_starvla][INFO] - Step 1352, Loss: {'train/loss': 0.0023439875803887844, 'train/grad_norm_pre_clip': 2.871366500854492, 'timing/data': 0.00034312717616558075, 'timing/model': 1.2323861769400537, 'learning_rate/action_model': 2.67916177618578e-05, 'learning_rate/action_query_proj': 2.67916177618578e-05, 'learning_rate/base': 6.69790444046445e-06, 'epoch': 0.58}) +[2026-06-30 04:36:50,298][starVLA.training.train_starvla][INFO] - Step 1353, Loss: {'train/loss': 0.35947471857070923, 'train/grad_norm_pre_clip': 2.297811508178711, 'timing/data': 0.0002224799245595932, 'timing/model': 1.2304459258448333, 'learning_rate/action_model': 2.6727485415538572e-05, 'learning_rate/action_query_proj': 2.6727485415538572e-05, 'learning_rate/base': 6.681871353884643e-06, 'epoch': 0.58}) +[2026-06-30 04:37:22,924][starVLA.training.train_starvla][INFO] - Step 1354, Loss: {'train/loss': 0.00011155915854033083, 'train/grad_norm_pre_clip': 1.872198462486267, 'timing/data': 0.0002023021224886179, 'timing/model': 1.2298761811107397, 'learning_rate/action_model': 2.6663415426636818e-05, 'learning_rate/action_query_proj': 2.6663415426636818e-05, 'learning_rate/base': 6.6658538566592046e-06, 'epoch': 0.58}) +[2026-06-30 04:37:55,622][starVLA.training.train_starvla][INFO] - Step 1355, Loss: {'train/loss': 0.0011137045221403241, 'train/grad_norm_pre_clip': 1.8450727462768555, 'timing/data': 0.00021977396681904793, 'timing/model': 1.2287172030191869, 'learning_rate/action_model': 2.659940795323886e-05, 'learning_rate/action_query_proj': 2.659940795323886e-05, 'learning_rate/base': 6.649851988309715e-06, 'epoch': 0.58}) +[2026-06-30 04:38:28,284][starVLA.training.train_starvla][INFO] - Step 1356, Loss: {'train/loss': 0.025598563253879547, 'train/grad_norm_pre_clip': 2.4574549198150635, 'timing/data': 0.0002594399265944958, 'timing/model': 1.2301003627944738, 'learning_rate/action_model': 2.6535463153276757e-05, 'learning_rate/action_query_proj': 2.6535463153276757e-05, 'learning_rate/base': 6.633865788319189e-06, 'epoch': 0.58}) +[2026-06-30 04:39:00,924][starVLA.training.train_starvla][INFO] - Step 1357, Loss: {'train/loss': 0.013806910254061222, 'train/grad_norm_pre_clip': 1.7663556337356567, 'timing/data': 0.00024094409309327602, 'timing/model': 1.2281151770148426, 'learning_rate/action_model': 2.6471581184528012e-05, 'learning_rate/action_query_proj': 2.6471581184528012e-05, 'learning_rate/base': 6.617895296132003e-06, 'epoch': 0.58}) +[2026-06-30 04:39:33,584][starVLA.training.train_starvla][INFO] - Step 1358, Loss: {'train/loss': 0.018154731020331383, 'train/grad_norm_pre_clip': 2.822448253631592, 'timing/data': 0.0002833451144397259, 'timing/model': 1.2312852169852704, 'learning_rate/action_model': 2.640776220461495e-05, 'learning_rate/action_query_proj': 2.640776220461495e-05, 'learning_rate/base': 6.601940551153738e-06, 'epoch': 0.58}) +[2026-06-30 04:40:06,241][starVLA.training.train_starvla][INFO] - Step 1359, Loss: {'train/loss': 0.3266863226890564, 'train/grad_norm_pre_clip': 3.101961374282837, 'timing/data': 0.000352304894477129, 'timing/model': 1.2322280048392713, 'learning_rate/action_model': 2.634400637100461e-05, 'learning_rate/action_query_proj': 2.634400637100461e-05, 'learning_rate/base': 6.586001592751152e-06, 'epoch': 0.58}) +[2026-06-30 04:40:38,913][starVLA.training.train_starvla][INFO] - Step 1360, Loss: {'train/loss': 0.028081901371479034, 'train/grad_norm_pre_clip': 3.40468430519104, 'timing/data': 0.00019493792206048965, 'timing/model': 1.231775752035901, 'learning_rate/action_model': 2.6280313841008146e-05, 'learning_rate/action_query_proj': 2.6280313841008146e-05, 'learning_rate/base': 6.5700784602520366e-06, 'epoch': 0.59}) +[2026-06-30 04:41:11,580][starVLA.training.train_starvla][INFO] - Step 1361, Loss: {'train/loss': 0.01596997119486332, 'train/grad_norm_pre_clip': 4.2302985191345215, 'timing/data': 0.00019974587485194206, 'timing/model': 1.2312003059778363, 'learning_rate/action_model': 2.6216684771780608e-05, 'learning_rate/action_query_proj': 2.6216684771780608e-05, 'learning_rate/base': 6.554171192945152e-06, 'epoch': 0.59}) +[2026-06-30 04:41:44,252][starVLA.training.train_starvla][INFO] - Step 1362, Loss: {'train/loss': 1.6003519704099745e-05, 'train/grad_norm_pre_clip': 2.897094964981079, 'timing/data': 0.00020159082487225533, 'timing/model': 1.2331582088954747, 'learning_rate/action_model': 2.6153119320320306e-05, 'learning_rate/action_query_proj': 2.6153119320320306e-05, 'learning_rate/base': 6.538279830080076e-06, 'epoch': 0.59}) +[2026-06-30 04:42:16,924][starVLA.training.train_starvla][INFO] - Step 1363, Loss: {'train/loss': 6.269690493354574e-05, 'train/grad_norm_pre_clip': 1.6671035289764404, 'timing/data': 0.00021355296485126019, 'timing/model': 1.2325550010427833, 'learning_rate/action_model': 2.6089617643468746e-05, 'learning_rate/action_query_proj': 2.6089617643468746e-05, 'learning_rate/base': 6.5224044108671865e-06, 'epoch': 0.59}) +[2026-06-30 04:42:49,596][starVLA.training.train_starvla][INFO] - Step 1364, Loss: {'train/loss': 0.02315639704465866, 'train/grad_norm_pre_clip': 3.4714834690093994, 'timing/data': 0.00020043808035552502, 'timing/model': 1.2285506159532815, 'learning_rate/action_model': 2.6026179897909965e-05, 'learning_rate/action_query_proj': 2.6026179897909965e-05, 'learning_rate/base': 6.506544974477491e-06, 'epoch': 0.59}) +[2026-06-30 04:43:22,292][starVLA.training.train_starvla][INFO] - Step 1365, Loss: {'train/loss': 1.9669039829750545e-05, 'train/grad_norm_pre_clip': 3.2302513122558594, 'timing/data': 0.00020934478379786015, 'timing/model': 1.2294505338650197, 'learning_rate/action_model': 2.5962806240170305e-05, 'learning_rate/action_query_proj': 2.5962806240170305e-05, 'learning_rate/base': 6.490701560042576e-06, 'epoch': 0.59}) +[2026-06-30 04:43:54,944][starVLA.training.train_starvla][INFO] - Step 1366, Loss: {'train/loss': 0.0003546452207956463, 'train/grad_norm_pre_clip': 2.101769208908081, 'timing/data': 0.00030111800879240036, 'timing/model': 1.2263904311694205, 'learning_rate/action_model': 2.5899496826617953e-05, 'learning_rate/action_query_proj': 2.5899496826617953e-05, 'learning_rate/base': 6.474874206654488e-06, 'epoch': 0.59}) +[2026-06-30 04:44:27,643][starVLA.training.train_starvla][INFO] - Step 1367, Loss: {'train/loss': 6.586288236576365e-06, 'train/grad_norm_pre_clip': 2.9707298278808594, 'timing/data': 0.0002050371840596199, 'timing/model': 1.235303012887016, 'learning_rate/action_model': 2.5836251813462624e-05, 'learning_rate/action_query_proj': 2.5836251813462624e-05, 'learning_rate/base': 6.459062953365656e-06, 'epoch': 0.59}) +[2026-06-30 04:45:00,297][starVLA.training.train_starvla][INFO] - Step 1368, Loss: {'train/loss': 8.00372363300994e-05, 'train/grad_norm_pre_clip': 2.025125503540039, 'timing/data': 0.00025592301972210407, 'timing/model': 1.2307153700385243, 'learning_rate/action_model': 2.5773071356755085e-05, 'learning_rate/action_query_proj': 2.5773071356755085e-05, 'learning_rate/base': 6.443267839188771e-06, 'epoch': 0.59}) +[2026-06-30 04:45:32,976][starVLA.training.train_starvla][INFO] - Step 1369, Loss: {'train/loss': 3.45992339134682e-05, 'train/grad_norm_pre_clip': 2.1501142978668213, 'timing/data': 0.00021104887127876282, 'timing/model': 1.2216143801342696, 'learning_rate/action_model': 2.5709955612386833e-05, 'learning_rate/action_query_proj': 2.5709955612386833e-05, 'learning_rate/base': 6.427488903096708e-06, 'epoch': 0.59}) +[2026-06-30 04:46:05,619][starVLA.training.train_starvla][INFO] - Step 1370, Loss: {'train/loss': 0.3661297559738159, 'train/grad_norm_pre_clip': 1.2615070343017578, 'timing/data': 0.0002085540909320116, 'timing/model': 1.2300509260967374, 'learning_rate/action_model': 2.5646904736089667e-05, 'learning_rate/action_query_proj': 2.5646904736089667e-05, 'learning_rate/base': 6.411726184022417e-06, 'epoch': 0.59}) +[2026-06-30 04:46:38,255][starVLA.training.train_starvla][INFO] - Step 1371, Loss: {'train/loss': 0.011030998080968857, 'train/grad_norm_pre_clip': 1.1826272010803223, 'timing/data': 0.0003495090641081333, 'timing/model': 1.2325804980937392, 'learning_rate/action_model': 2.5583918883435405e-05, 'learning_rate/action_query_proj': 2.5583918883435405e-05, 'learning_rate/base': 6.395979720858851e-06, 'epoch': 0.59}) +[2026-06-30 04:47:10,922][starVLA.training.train_starvla][INFO] - Step 1372, Loss: {'train/loss': 0.0015451549552381039, 'train/grad_norm_pre_clip': 1.5328360795974731, 'timing/data': 0.0003168168477714062, 'timing/model': 1.2309059179387987, 'learning_rate/action_model': 2.552099820983535e-05, 'learning_rate/action_query_proj': 2.552099820983535e-05, 'learning_rate/base': 6.380249552458837e-06, 'epoch': 0.59}) +[2026-06-30 04:47:43,619][starVLA.training.train_starvla][INFO] - Step 1373, Loss: {'train/loss': 0.10069528967142105, 'train/grad_norm_pre_clip': 1.9320571422576904, 'timing/data': 0.0003000060096383095, 'timing/model': 1.2285496129188687, 'learning_rate/action_model': 2.5458142870540013e-05, 'learning_rate/action_query_proj': 2.5458142870540013e-05, 'learning_rate/base': 6.364535717635003e-06, 'epoch': 0.59}) +[2026-06-30 04:48:16,203][starVLA.training.train_starvla][INFO] - Step 1374, Loss: {'train/loss': 0.5590572953224182, 'train/grad_norm_pre_clip': 3.347698926925659, 'timing/data': 0.0001966410782188177, 'timing/model': 1.2332778330892324, 'learning_rate/action_model': 2.539535302063868e-05, 'learning_rate/action_query_proj': 2.539535302063868e-05, 'learning_rate/base': 6.34883825515967e-06, 'epoch': 0.59}) +[2026-06-30 04:48:48,885][starVLA.training.train_starvla][INFO] - Step 1375, Loss: {'train/loss': 0.37610113620758057, 'train/grad_norm_pre_clip': 2.6148979663848877, 'timing/data': 0.0002852680627256632, 'timing/model': 1.233946454944089, 'learning_rate/action_model': 2.5332628815059105e-05, 'learning_rate/action_query_proj': 2.5332628815059105e-05, 'learning_rate/base': 6.333157203764776e-06, 'epoch': 0.59}) +[2026-06-30 04:49:21,530][starVLA.training.train_starvla][INFO] - Step 1376, Loss: {'train/loss': 0.0547814816236496, 'train/grad_norm_pre_clip': 1.9697185754776, 'timing/data': 0.0002572860103100538, 'timing/model': 1.2315395439509302, 'learning_rate/action_model': 2.5269970408566994e-05, 'learning_rate/action_query_proj': 2.5269970408566994e-05, 'learning_rate/base': 6.3174926021417486e-06, 'epoch': 0.59}) +[2026-06-30 04:49:54,218][starVLA.training.train_starvla][INFO] - Step 1377, Loss: {'train/loss': 0.027510756626725197, 'train/grad_norm_pre_clip': 3.8187856674194336, 'timing/data': 0.0003320269752293825, 'timing/model': 1.2297248500399292, 'learning_rate/action_model': 2.5207377955765722e-05, 'learning_rate/action_query_proj': 2.5207377955765722e-05, 'learning_rate/base': 6.3018444889414305e-06, 'epoch': 0.59}) +[2026-06-30 04:50:26,917][starVLA.training.train_starvla][INFO] - Step 1378, Loss: {'train/loss': 0.04920655116438866, 'train/grad_norm_pre_clip': 2.481886625289917, 'timing/data': 0.00025655399076640606, 'timing/model': 1.230705930152908, 'learning_rate/action_model': 2.514485161109599e-05, 'learning_rate/action_query_proj': 2.514485161109599e-05, 'learning_rate/base': 6.286212902773997e-06, 'epoch': 0.59}) +[2026-06-30 04:50:59,613][starVLA.training.train_starvla][INFO] - Step 1379, Loss: {'train/loss': 0.03857119753956795, 'train/grad_norm_pre_clip': 2.0076088905334473, 'timing/data': 0.000374015886336565, 'timing/model': 1.231450866209343, 'learning_rate/action_model': 2.5082391528835318e-05, 'learning_rate/action_query_proj': 2.5082391528835318e-05, 'learning_rate/base': 6.2705978822088295e-06, 'epoch': 0.59}) +[2026-06-30 04:51:32,307][starVLA.training.train_starvla][INFO] - Step 1380, Loss: {'train/loss': 0.02747957780957222, 'train/grad_norm_pre_clip': 2.3264520168304443, 'timing/data': 0.00020193099044263363, 'timing/model': 1.221700371010229, 'learning_rate/action_model': 2.5019997863097743e-05, 'learning_rate/action_query_proj': 2.5019997863097743e-05, 'learning_rate/base': 6.254999465774436e-06, 'epoch': 0.59}) +[2026-06-30 04:52:05,023][starVLA.training.train_starvla][INFO] - Step 1381, Loss: {'train/loss': 0.01793574169278145, 'train/grad_norm_pre_clip': 1.7169145345687866, 'timing/data': 0.00019513815641403198, 'timing/model': 1.2304660778027028, 'learning_rate/action_model': 2.495767076783343e-05, 'learning_rate/action_query_proj': 2.495767076783343e-05, 'learning_rate/base': 6.2394176919583575e-06, 'epoch': 0.59}) +[2026-06-30 04:52:37,713][starVLA.training.train_starvla][INFO] - Step 1382, Loss: {'train/loss': 0.020596057176589966, 'train/grad_norm_pre_clip': 1.055850625038147, 'timing/data': 0.00030518602579832077, 'timing/model': 1.2340361229144037, 'learning_rate/action_model': 2.4895410396828305e-05, 'learning_rate/action_query_proj': 2.4895410396828305e-05, 'learning_rate/base': 6.223852599207076e-06, 'epoch': 0.59}) +[2026-06-30 04:53:10,418][starVLA.training.train_starvla][INFO] - Step 1383, Loss: {'train/loss': 0.0001211204071296379, 'train/grad_norm_pre_clip': 2.069859027862549, 'timing/data': 0.00021947408095002174, 'timing/model': 1.2268468600232154, 'learning_rate/action_model': 2.4833216903703644e-05, 'learning_rate/action_query_proj': 2.4833216903703644e-05, 'learning_rate/base': 6.208304225925911e-06, 'epoch': 0.6}) +[2026-06-30 04:53:43,122][starVLA.training.train_starvla][INFO] - Step 1384, Loss: {'train/loss': 0.0825064554810524, 'train/grad_norm_pre_clip': 1.3837970495224, 'timing/data': 0.00020223204046487808, 'timing/model': 1.2195304268971086, 'learning_rate/action_model': 2.4771090441915696e-05, 'learning_rate/action_query_proj': 2.4771090441915696e-05, 'learning_rate/base': 6.192772610478924e-06, 'epoch': 0.6}) +[2026-06-30 04:54:15,852][starVLA.training.train_starvla][INFO] - Step 1385, Loss: {'train/loss': 0.6874056458473206, 'train/grad_norm_pre_clip': 3.0978949069976807, 'timing/data': 0.00019312510266900063, 'timing/model': 1.2371601059567183, 'learning_rate/action_model': 2.4709031164755326e-05, 'learning_rate/action_query_proj': 2.4709031164755326e-05, 'learning_rate/base': 6.177257791188832e-06, 'epoch': 0.6}) +[2026-06-30 04:54:48,487][starVLA.training.train_starvla][INFO] - Step 1386, Loss: {'train/loss': 0.20380231738090515, 'train/grad_norm_pre_clip': 2.674980401992798, 'timing/data': 0.00019696098752319813, 'timing/model': 1.233665914973244, 'learning_rate/action_model': 2.4647039225347657e-05, 'learning_rate/action_query_proj': 2.4647039225347657e-05, 'learning_rate/base': 6.161759806336914e-06, 'epoch': 0.6}) +[2026-06-30 04:55:21,179][starVLA.training.train_starvla][INFO] - Step 1387, Loss: {'train/loss': 0.005252944305539131, 'train/grad_norm_pre_clip': 2.7552993297576904, 'timing/data': 0.00025472091510891914, 'timing/model': 1.2334485640749335, 'learning_rate/action_model': 2.458511477665163e-05, 'learning_rate/action_query_proj': 2.458511477665163e-05, 'learning_rate/base': 6.146278694162907e-06, 'epoch': 0.6}) +[2026-06-30 04:55:53,888][starVLA.training.train_starvla][INFO] - Step 1388, Loss: {'train/loss': 6.496878995676525e-06, 'train/grad_norm_pre_clip': 3.3163869380950928, 'timing/data': 0.00020423484966158867, 'timing/model': 1.2348031690344214, 'learning_rate/action_model': 2.4523257971459646e-05, 'learning_rate/action_query_proj': 2.4523257971459646e-05, 'learning_rate/base': 6.1308144928649116e-06, 'epoch': 0.6}) +[2026-06-30 04:56:26,526][starVLA.training.train_starvla][INFO] - Step 1389, Loss: {'train/loss': 1.0371135431341827e-05, 'train/grad_norm_pre_clip': 2.998887300491333, 'timing/data': 0.00022692792117595673, 'timing/model': 1.2317631638143212, 'learning_rate/action_model': 2.446146896239722e-05, 'learning_rate/action_query_proj': 2.446146896239722e-05, 'learning_rate/base': 6.115367240599305e-06, 'epoch': 0.6}) +[2026-06-30 04:56:59,186][starVLA.training.train_starvla][INFO] - Step 1390, Loss: {'train/loss': 0.05573089420795441, 'train/grad_norm_pre_clip': 1.7206827402114868, 'timing/data': 0.0002865500282496214, 'timing/model': 1.231476651970297, 'learning_rate/action_model': 2.4399747901922642e-05, 'learning_rate/action_query_proj': 2.4399747901922642e-05, 'learning_rate/base': 6.0999369754806605e-06, 'epoch': 0.6}) +[2026-06-30 04:57:31,868][starVLA.training.train_starvla][INFO] - Step 1391, Loss: {'train/loss': 0.19985368847846985, 'train/grad_norm_pre_clip': 2.8899898529052734, 'timing/data': 0.00019576889462769032, 'timing/model': 1.235880307853222, 'learning_rate/action_model': 2.4338094942326424e-05, 'learning_rate/action_query_proj': 2.4338094942326424e-05, 'learning_rate/base': 6.084523735581606e-06, 'epoch': 0.6}) +[2026-06-30 04:58:04,534][starVLA.training.train_starvla][INFO] - Step 1392, Loss: {'train/loss': 0.0091498838737607, 'train/grad_norm_pre_clip': 2.847165107727051, 'timing/data': 0.00019339402206242085, 'timing/model': 1.227321689017117, 'learning_rate/action_model': 2.4276510235731164e-05, 'learning_rate/action_query_proj': 2.4276510235731164e-05, 'learning_rate/base': 6.069127558932791e-06, 'epoch': 0.6}) +[2026-06-30 04:58:37,197][starVLA.training.train_starvla][INFO] - Step 1393, Loss: {'train/loss': 0.0016909432597458363, 'train/grad_norm_pre_clip': 1.3806384801864624, 'timing/data': 0.00025116303004324436, 'timing/model': 1.22892662207596, 'learning_rate/action_model': 2.421499393409097e-05, 'learning_rate/action_query_proj': 2.421499393409097e-05, 'learning_rate/base': 6.053748483522742e-06, 'epoch': 0.6}) +[2026-06-30 04:59:09,847][starVLA.training.train_starvla][INFO] - Step 1394, Loss: {'train/loss': 0.09277995675802231, 'train/grad_norm_pre_clip': 1.1563198566436768, 'timing/data': 0.0003404109738767147, 'timing/model': 1.2328069240320474, 'learning_rate/action_model': 2.4153546189191244e-05, 'learning_rate/action_query_proj': 2.4153546189191244e-05, 'learning_rate/base': 6.038386547297811e-06, 'epoch': 0.6}) +[2026-06-30 04:59:42,552][starVLA.training.train_starvla][INFO] - Step 1395, Loss: {'train/loss': 0.08201282471418381, 'train/grad_norm_pre_clip': 2.644059181213379, 'timing/data': 0.00032642483711242676, 'timing/model': 1.2296309450175613, 'learning_rate/action_model': 2.4092167152648133e-05, 'learning_rate/action_query_proj': 2.4092167152648133e-05, 'learning_rate/base': 6.023041788162033e-06, 'epoch': 0.6}) +[2026-06-30 05:00:15,234][starVLA.training.train_starvla][INFO] - Step 1396, Loss: {'train/loss': 0.002231090096756816, 'train/grad_norm_pre_clip': 2.2985355854034424, 'timing/data': 0.00025936891324818134, 'timing/model': 1.2320745149627328, 'learning_rate/action_model': 2.4030856975908357e-05, 'learning_rate/action_query_proj': 2.4030856975908357e-05, 'learning_rate/base': 6.007714243977089e-06, 'epoch': 0.6}) +[2026-06-30 05:00:47,909][starVLA.training.train_starvla][INFO] - Step 1397, Loss: {'train/loss': 9.965058416128159e-05, 'train/grad_norm_pre_clip': 1.9814471006393433, 'timing/data': 0.00021363189443945885, 'timing/model': 1.230027220910415, 'learning_rate/action_model': 2.3969615810248657e-05, 'learning_rate/action_query_proj': 2.3969615810248657e-05, 'learning_rate/base': 5.992403952562164e-06, 'epoch': 0.6}) +[2026-06-30 05:01:20,606][starVLA.training.train_starvla][INFO] - Step 1398, Loss: {'train/loss': 0.4880683422088623, 'train/grad_norm_pre_clip': 3.9066169261932373, 'timing/data': 0.0002966888714581728, 'timing/model': 1.2270606248639524, 'learning_rate/action_model': 2.3908443806775532e-05, 'learning_rate/action_query_proj': 2.3908443806775532e-05, 'learning_rate/base': 5.977110951693883e-06, 'epoch': 0.6}) +[2026-06-30 05:01:53,257][starVLA.training.train_starvla][INFO] - Step 1399, Loss: {'train/loss': 7.706166070420295e-05, 'train/grad_norm_pre_clip': 3.3566675186157227, 'timing/data': 0.00021655787713825703, 'timing/model': 1.2306196079589427, 'learning_rate/action_model': 2.3847341116424787e-05, 'learning_rate/action_query_proj': 2.3847341116424787e-05, 'learning_rate/base': 5.961835279106197e-06, 'epoch': 0.6}) +[2026-06-30 05:02:25,965][starVLA.training.train_starvla][INFO] - Step 1400, Loss: {'train/loss': 0.017248038202524185, 'train/grad_norm_pre_clip': 1.2797505855560303, 'timing/data': 0.00020284205675125122, 'timing/model': 1.234666720032692, 'learning_rate/action_model': 2.3786307889961294e-05, 'learning_rate/action_query_proj': 2.3786307889961294e-05, 'learning_rate/base': 5.9465769724903235e-06, 'epoch': 0.6}) +[2026-06-30 05:02:58,660][starVLA.training.train_starvla][INFO] - Step 1401, Loss: {'train/loss': 0.00029925000853836536, 'train/grad_norm_pre_clip': 2.026449203491211, 'timing/data': 0.0001989749725908041, 'timing/model': 1.233298190170899, 'learning_rate/action_model': 2.372534427797844e-05, 'learning_rate/action_query_proj': 2.372534427797844e-05, 'learning_rate/base': 5.93133606949461e-06, 'epoch': 0.6}) +[2026-06-30 05:03:31,300][starVLA.training.train_starvla][INFO] - Step 1402, Loss: {'train/loss': 0.003899375908076763, 'train/grad_norm_pre_clip': 2.2495737075805664, 'timing/data': 0.00019776285625994205, 'timing/model': 1.2304960088804364, 'learning_rate/action_model': 2.366445043089787e-05, 'learning_rate/action_query_proj': 2.366445043089787e-05, 'learning_rate/base': 5.916112607724468e-06, 'epoch': 0.6}) +[2026-06-30 05:04:04,173][starVLA.training.train_starvla][INFO] - Step 1403, Loss: {'train/loss': 0.11495475471019745, 'train/grad_norm_pre_clip': 1.5713865756988525, 'timing/data': 0.0002796468324959278, 'timing/model': 1.229650504887104, 'learning_rate/action_model': 2.36036264989691e-05, 'learning_rate/action_query_proj': 2.36036264989691e-05, 'learning_rate/base': 5.900906624742275e-06, 'epoch': 0.6}) +[2026-06-30 05:04:36,775][starVLA.training.train_starvla][INFO] - Step 1404, Loss: {'train/loss': 0.007840847596526146, 'train/grad_norm_pre_clip': 2.5858497619628906, 'timing/data': 0.00030197901651263237, 'timing/model': 1.2275489738676697, 'learning_rate/action_model': 2.3542872632269158e-05, 'learning_rate/action_query_proj': 2.3542872632269158e-05, 'learning_rate/base': 5.885718158067289e-06, 'epoch': 0.6}) +[2026-06-30 05:05:09,442][starVLA.training.train_starvla][INFO] - Step 1405, Loss: {'train/loss': 0.12458407133817673, 'train/grad_norm_pre_clip': 1.7734593152999878, 'timing/data': 0.0001980629749596119, 'timing/model': 1.2307415730319917, 'learning_rate/action_model': 2.3482188980702154e-05, 'learning_rate/action_query_proj': 2.3482188980702154e-05, 'learning_rate/base': 5.870547245175538e-06, 'epoch': 0.6}) +[2026-06-30 05:05:42,135][starVLA.training.train_starvla][INFO] - Step 1406, Loss: {'train/loss': 6.824703632446472e-06, 'train/grad_norm_pre_clip': 3.02813458442688, 'timing/data': 0.00036628986708819866, 'timing/model': 1.2276387601159513, 'learning_rate/action_model': 2.342157569399897e-05, 'learning_rate/action_query_proj': 2.342157569399897e-05, 'learning_rate/base': 5.855393923499743e-06, 'epoch': 0.6}) +[2026-06-30 05:06:14,800][starVLA.training.train_starvla][INFO] - Step 1407, Loss: {'train/loss': 0.2851174473762512, 'train/grad_norm_pre_clip': 1.3050082921981812, 'timing/data': 0.0002544790040701628, 'timing/model': 1.229621654143557, 'learning_rate/action_model': 2.3361032921716848e-05, 'learning_rate/action_query_proj': 2.3361032921716848e-05, 'learning_rate/base': 5.840258230429212e-06, 'epoch': 0.61}) +[2026-06-30 05:06:47,459][starVLA.training.train_starvla][INFO] - Step 1408, Loss: {'train/loss': 0.0023520817048847675, 'train/grad_norm_pre_clip': 2.1013593673706055, 'timing/data': 0.0003615620080381632, 'timing/model': 1.2327981719281524, 'learning_rate/action_model': 2.3300560813239113e-05, 'learning_rate/action_query_proj': 2.3300560813239113e-05, 'learning_rate/base': 5.825140203309778e-06, 'epoch': 0.61}) +[2026-06-30 05:07:20,127][starVLA.training.train_starvla][INFO] - Step 1409, Loss: {'train/loss': 0.11885643750429153, 'train/grad_norm_pre_clip': 2.3834633827209473, 'timing/data': 0.0001937949564307928, 'timing/model': 1.2328490719664842, 'learning_rate/action_model': 2.32401595177746e-05, 'learning_rate/action_query_proj': 2.32401595177746e-05, 'learning_rate/base': 5.81003987944365e-06, 'epoch': 0.61}) +[2026-06-30 05:07:52,798][starVLA.training.train_starvla][INFO] - Step 1410, Loss: {'train/loss': 0.00016552276792936027, 'train/grad_norm_pre_clip': 1.9147757291793823, 'timing/data': 0.0002140828873962164, 'timing/model': 1.2294060809072107, 'learning_rate/action_model': 2.3179829184357572e-05, 'learning_rate/action_query_proj': 2.3179829184357572e-05, 'learning_rate/base': 5.794957296089393e-06, 'epoch': 0.61}) +[2026-06-30 05:08:25,482][starVLA.training.train_starvla][INFO] - Step 1411, Loss: {'train/loss': 0.8771125078201294, 'train/grad_norm_pre_clip': 2.710207223892212, 'timing/data': 0.00020499597303569317, 'timing/model': 1.2308483531232923, 'learning_rate/action_model': 2.311956996184708e-05, 'learning_rate/action_query_proj': 2.311956996184708e-05, 'learning_rate/base': 5.77989249046177e-06, 'epoch': 0.61}) +[2026-06-30 05:08:58,119][starVLA.training.train_starvla][INFO] - Step 1412, Loss: {'train/loss': 0.3836289048194885, 'train/grad_norm_pre_clip': 2.7328178882598877, 'timing/data': 0.00019795307889580727, 'timing/model': 1.2315171419177204, 'learning_rate/action_model': 2.3059381998926825e-05, 'learning_rate/action_query_proj': 2.3059381998926825e-05, 'learning_rate/base': 5.764845499731706e-06, 'epoch': 0.61}) +[2026-06-30 05:09:30,765][starVLA.training.train_starvla][INFO] - Step 1413, Loss: {'train/loss': 0.0017152895452454686, 'train/grad_norm_pre_clip': 2.815192937850952, 'timing/data': 0.00019916496239602566, 'timing/model': 1.237551188096404, 'learning_rate/action_model': 2.299926544410455e-05, 'learning_rate/action_query_proj': 2.299926544410455e-05, 'learning_rate/base': 5.749816361026138e-06, 'epoch': 0.61}) +[2026-06-30 05:10:03,425][starVLA.training.train_starvla][INFO] - Step 1414, Loss: {'train/loss': 0.5496953129768372, 'train/grad_norm_pre_clip': 3.2286899089813232, 'timing/data': 0.00027753296308219433, 'timing/model': 1.232792248018086, 'learning_rate/action_model': 2.293922044571194e-05, 'learning_rate/action_query_proj': 2.293922044571194e-05, 'learning_rate/base': 5.734805111427985e-06, 'epoch': 0.61}) +[2026-06-30 05:10:36,061][starVLA.training.train_starvla][INFO] - Step 1415, Loss: {'train/loss': 0.001159827341325581, 'train/grad_norm_pre_clip': 2.6167943477630615, 'timing/data': 0.000198825029656291, 'timing/model': 1.2287903781980276, 'learning_rate/action_model': 2.287924715190402e-05, 'learning_rate/action_query_proj': 2.287924715190402e-05, 'learning_rate/base': 5.719811787976005e-06, 'epoch': 0.61}) +[2026-06-30 05:11:08,759][starVLA.training.train_starvla][INFO] - Step 1416, Loss: {'train/loss': 0.002802510978654027, 'train/grad_norm_pre_clip': 2.8886818885803223, 'timing/data': 0.00024698488414287567, 'timing/model': 1.2328633659053594, 'learning_rate/action_model': 2.2819345710658952e-05, 'learning_rate/action_query_proj': 2.2819345710658952e-05, 'learning_rate/base': 5.704836427664738e-06, 'epoch': 0.61}) +[2026-06-30 05:11:41,414][starVLA.training.train_starvla][INFO] - Step 1417, Loss: {'train/loss': 0.0011060840915888548, 'train/grad_norm_pre_clip': 2.715383768081665, 'timing/data': 0.00018688198179006577, 'timing/model': 1.229915845906362, 'learning_rate/action_model': 2.2759516269777563e-05, 'learning_rate/action_query_proj': 2.2759516269777563e-05, 'learning_rate/base': 5.689879067444391e-06, 'epoch': 0.61}) +[2026-06-30 05:12:14,089][starVLA.training.train_starvla][INFO] - Step 1418, Loss: {'train/loss': 0.317778080701828, 'train/grad_norm_pre_clip': 1.3103811740875244, 'timing/data': 0.00020666001364588737, 'timing/model': 1.2337581631727517, 'learning_rate/action_model': 2.2699758976883085e-05, 'learning_rate/action_query_proj': 2.2699758976883085e-05, 'learning_rate/base': 5.674939744220771e-06, 'epoch': 0.61}) +[2026-06-30 05:12:46,792][starVLA.training.train_starvla][INFO] - Step 1419, Loss: {'train/loss': 0.0035898557398468256, 'train/grad_norm_pre_clip': 2.808900833129883, 'timing/data': 0.00019787298515439034, 'timing/model': 1.2273701198864728, 'learning_rate/action_model': 2.2640073979420673e-05, 'learning_rate/action_query_proj': 2.2640073979420673e-05, 'learning_rate/base': 5.660018494855168e-06, 'epoch': 0.61}) +[2026-06-30 05:13:19,514][starVLA.training.train_starvla][INFO] - Step 1420, Loss: {'train/loss': 0.0011569568887352943, 'train/grad_norm_pre_clip': 2.1311466693878174, 'timing/data': 0.00019349507056176662, 'timing/model': 1.232742756139487, 'learning_rate/action_model': 2.258046142465714e-05, 'learning_rate/action_query_proj': 2.258046142465714e-05, 'learning_rate/base': 5.645115356164285e-06, 'epoch': 0.61}) +[2026-06-30 05:13:52,185][starVLA.training.train_starvla][INFO] - Step 1421, Loss: {'train/loss': 7.124772673705593e-05, 'train/grad_norm_pre_clip': 1.2965054512023926, 'timing/data': 0.00020110886543989182, 'timing/model': 1.2311535240150988, 'learning_rate/action_model': 2.2520921459680512e-05, 'learning_rate/action_query_proj': 2.2520921459680512e-05, 'learning_rate/base': 5.630230364920128e-06, 'epoch': 0.61}) +[2026-06-30 05:14:24,903][starVLA.training.train_starvla][INFO] - Step 1422, Loss: {'train/loss': 1.1503551831992809e-05, 'train/grad_norm_pre_clip': 2.5613820552825928, 'timing/data': 0.00019955681636929512, 'timing/model': 1.2324176980182528, 'learning_rate/action_model': 2.2461454231399765e-05, 'learning_rate/action_query_proj': 2.2461454231399765e-05, 'learning_rate/base': 5.615363557849941e-06, 'epoch': 0.61}) +[2026-06-30 05:14:57,589][starVLA.training.train_starvla][INFO] - Step 1423, Loss: {'train/loss': 0.021640708670020103, 'train/grad_norm_pre_clip': 3.3958792686462402, 'timing/data': 0.0002417250070720911, 'timing/model': 1.2335256959777325, 'learning_rate/action_model': 2.240205988654437e-05, 'learning_rate/action_query_proj': 2.240205988654437e-05, 'learning_rate/base': 5.600514971636092e-06, 'epoch': 0.61}) +[2026-06-30 05:15:30,259][starVLA.training.train_starvla][INFO] - Step 1424, Loss: {'train/loss': 0.0645805299282074, 'train/grad_norm_pre_clip': 2.2301342487335205, 'timing/data': 0.0002478270325809717, 'timing/model': 1.2333031930029392, 'learning_rate/action_model': 2.234273857166396e-05, 'learning_rate/action_query_proj': 2.234273857166396e-05, 'learning_rate/base': 5.58568464291599e-06, 'epoch': 0.61}) +[2026-06-30 05:16:02,924][starVLA.training.train_starvla][INFO] - Step 1425, Loss: {'train/loss': 0.0010653450153768063, 'train/grad_norm_pre_clip': 3.9099788665771484, 'timing/data': 0.00030633690766990185, 'timing/model': 1.2348388731479645, 'learning_rate/action_model': 2.2283490433127963e-05, 'learning_rate/action_query_proj': 2.2283490433127963e-05, 'learning_rate/base': 5.570872608281991e-06, 'epoch': 0.61}) +[2026-06-30 05:16:35,592][starVLA.training.train_starvla][INFO] - Step 1426, Loss: {'train/loss': 0.14654946327209473, 'train/grad_norm_pre_clip': 2.4573171138763428, 'timing/data': 0.0004445877857506275, 'timing/model': 1.2295418670400977, 'learning_rate/action_model': 2.222431561712533e-05, 'learning_rate/action_query_proj': 2.222431561712533e-05, 'learning_rate/base': 5.5560789042813324e-06, 'epoch': 0.61}) +[2026-06-30 05:17:08,284][starVLA.training.train_starvla][INFO] - Step 1427, Loss: {'train/loss': 0.0003881888696923852, 'train/grad_norm_pre_clip': 1.9244236946105957, 'timing/data': 0.00020821299403905869, 'timing/model': 1.234386342111975, 'learning_rate/action_model': 2.2165214269663958e-05, 'learning_rate/action_query_proj': 2.2165214269663958e-05, 'learning_rate/base': 5.5413035674159894e-06, 'epoch': 0.61}) +[2026-06-30 05:17:40,976][starVLA.training.train_starvla][INFO] - Step 1428, Loss: {'train/loss': 0.04214302450418472, 'train/grad_norm_pre_clip': 4.940196990966797, 'timing/data': 0.0002511739730834961, 'timing/model': 1.2354029170237482, 'learning_rate/action_model': 2.2106186536570605e-05, 'learning_rate/action_query_proj': 2.2106186536570605e-05, 'learning_rate/base': 5.526546634142651e-06, 'epoch': 0.61}) +[2026-06-30 05:18:13,643][starVLA.training.train_starvla][INFO] - Step 1429, Loss: {'train/loss': 0.00023363543732557446, 'train/grad_norm_pre_clip': 1.9633482694625854, 'timing/data': 0.00033236597664654255, 'timing/model': 1.2318208578508347, 'learning_rate/action_model': 2.2047232563490307e-05, 'learning_rate/action_query_proj': 2.2047232563490307e-05, 'learning_rate/base': 5.511808140872577e-06, 'epoch': 0.61}) +[2026-06-30 05:18:46,343][starVLA.training.train_starvla][INFO] - Step 1430, Loss: {'train/loss': 0.100786954164505, 'train/grad_norm_pre_clip': 3.751835584640503, 'timing/data': 0.000355269992724061, 'timing/model': 1.2325611892156303, 'learning_rate/action_model': 2.1988352495886165e-05, 'learning_rate/action_query_proj': 2.1988352495886165e-05, 'learning_rate/base': 5.497088123971541e-06, 'epoch': 0.62}) +[2026-06-30 05:19:19,029][starVLA.training.train_starvla][INFO] - Step 1431, Loss: {'train/loss': 5.1702842029044405e-05, 'train/grad_norm_pre_clip': 2.9663290977478027, 'timing/data': 0.00019760197028517723, 'timing/model': 1.2291963470634073, 'learning_rate/action_model': 2.1929546479038843e-05, 'learning_rate/action_query_proj': 2.1929546479038843e-05, 'learning_rate/base': 5.482386619759711e-06, 'epoch': 0.62}) +[2026-06-30 05:19:51,704][starVLA.training.train_starvla][INFO] - Step 1432, Loss: {'train/loss': 0.1968824863433838, 'train/grad_norm_pre_clip': 2.4737470149993896, 'timing/data': 0.00038099801167845726, 'timing/model': 1.2309696828015149, 'learning_rate/action_model': 2.1870814658046384e-05, 'learning_rate/action_query_proj': 2.1870814658046384e-05, 'learning_rate/base': 5.467703664511596e-06, 'epoch': 0.62}) +[2026-06-30 05:20:24,344][starVLA.training.train_starvla][INFO] - Step 1433, Loss: {'train/loss': 0.18088752031326294, 'train/grad_norm_pre_clip': 2.665026903152466, 'timing/data': 0.00027190311811864376, 'timing/model': 1.2314904460217804, 'learning_rate/action_model': 2.18121571778237e-05, 'learning_rate/action_query_proj': 2.18121571778237e-05, 'learning_rate/base': 5.453039294455925e-06, 'epoch': 0.62}) +[2026-06-30 05:20:57,049][starVLA.training.train_starvla][INFO] - Step 1434, Loss: {'train/loss': 8.519450057065114e-05, 'train/grad_norm_pre_clip': 2.38598370552063, 'timing/data': 0.0003270660527050495, 'timing/model': 1.2305493098683655, 'learning_rate/action_model': 2.1753574183102292e-05, 'learning_rate/action_query_proj': 2.1753574183102292e-05, 'learning_rate/base': 5.438393545775573e-06, 'epoch': 0.62}) +[2026-06-30 05:21:29,785][starVLA.training.train_starvla][INFO] - Step 1435, Loss: {'train/loss': 3.0696342037117574e-06, 'train/grad_norm_pre_clip': 1.3539024591445923, 'timing/data': 0.0002996050752699375, 'timing/model': 1.2331370480824262, 'learning_rate/action_model': 2.169506581842987e-05, 'learning_rate/action_query_proj': 2.169506581842987e-05, 'learning_rate/base': 5.423766454607468e-06, 'epoch': 0.62}) +[2026-06-30 05:22:02,466][starVLA.training.train_starvla][INFO] - Step 1436, Loss: {'train/loss': 0.44007912278175354, 'train/grad_norm_pre_clip': 3.83853816986084, 'timing/data': 0.00020051817409694195, 'timing/model': 1.2309160670265555, 'learning_rate/action_model': 2.163663222817003e-05, 'learning_rate/action_query_proj': 2.163663222817003e-05, 'learning_rate/base': 5.409158057042507e-06, 'epoch': 0.62}) +[2026-06-30 05:22:35,056][starVLA.training.train_starvla][INFO] - Step 1437, Loss: {'train/loss': 0.3675622045993805, 'train/grad_norm_pre_clip': 3.5675997734069824, 'timing/data': 0.00020551797933876514, 'timing/model': 1.2196498268749565, 'learning_rate/action_model': 2.157827355650183e-05, 'learning_rate/action_query_proj': 2.157827355650183e-05, 'learning_rate/base': 5.394568389125458e-06, 'epoch': 0.62}) +[2026-06-30 05:23:07,705][starVLA.training.train_starvla][INFO] - Step 1438, Loss: {'train/loss': 0.026802165433764458, 'train/grad_norm_pre_clip': 1.861310362815857, 'timing/data': 0.0004472630098462105, 'timing/model': 1.2185038910247386, 'learning_rate/action_model': 2.15199899474195e-05, 'learning_rate/action_query_proj': 2.15199899474195e-05, 'learning_rate/base': 5.379997486854875e-06, 'epoch': 0.62}) +[2026-06-30 05:23:40,385][starVLA.training.train_starvla][INFO] - Step 1439, Loss: {'train/loss': 0.11712665110826492, 'train/grad_norm_pre_clip': 2.4776649475097656, 'timing/data': 0.0002672539558261633, 'timing/model': 1.2327587557956576, 'learning_rate/action_model': 2.1461781544732028e-05, 'learning_rate/action_query_proj': 2.1461781544732028e-05, 'learning_rate/base': 5.365445386183007e-06, 'epoch': 0.62}) +[2026-06-30 05:24:13,065][starVLA.training.train_starvla][INFO] - Step 1440, Loss: {'train/loss': 0.002745936391875148, 'train/grad_norm_pre_clip': 1.168853759765625, 'timing/data': 0.0002805190160870552, 'timing/model': 1.2296868870034814, 'learning_rate/action_model': 2.1403648492062894e-05, 'learning_rate/action_query_proj': 2.1403648492062894e-05, 'learning_rate/base': 5.350912123015723e-06, 'epoch': 0.62}) +[2026-06-30 05:24:45,726][starVLA.training.train_starvla][INFO] - Step 1441, Loss: {'train/loss': 0.164263054728508, 'train/grad_norm_pre_clip': 2.042351722717285, 'timing/data': 0.00020439503714442253, 'timing/model': 1.2270344467833638, 'learning_rate/action_model': 2.1345590932849613e-05, 'learning_rate/action_query_proj': 2.1345590932849613e-05, 'learning_rate/base': 5.336397733212403e-06, 'epoch': 0.62}) +[2026-06-30 05:25:18,429][starVLA.training.train_starvla][INFO] - Step 1442, Loss: {'train/loss': 0.0001116168568842113, 'train/grad_norm_pre_clip': 1.1753227710723877, 'timing/data': 0.000289425952360034, 'timing/model': 1.2352205449715257, 'learning_rate/action_model': 2.1287609010343445e-05, 'learning_rate/action_query_proj': 2.1287609010343445e-05, 'learning_rate/base': 5.321902252585861e-06, 'epoch': 0.62}) +[2026-06-30 05:25:51,116][starVLA.training.train_starvla][INFO] - Step 1443, Loss: {'train/loss': 0.009352610446512699, 'train/grad_norm_pre_clip': 3.159230947494507, 'timing/data': 0.0003171381540596485, 'timing/model': 1.2319582039490342, 'learning_rate/action_model': 2.122970286760899e-05, 'learning_rate/action_query_proj': 2.122970286760899e-05, 'learning_rate/base': 5.3074257169022475e-06, 'epoch': 0.62}) +[2026-06-30 05:26:23,814][starVLA.training.train_starvla][INFO] - Step 1444, Loss: {'train/loss': 0.1737818568944931, 'train/grad_norm_pre_clip': 2.3881404399871826, 'timing/data': 0.0002473869826644659, 'timing/model': 1.2279538330622017, 'learning_rate/action_model': 2.117187264752395e-05, 'learning_rate/action_query_proj': 2.117187264752395e-05, 'learning_rate/base': 5.292968161880987e-06, 'epoch': 0.62}) +[2026-06-30 05:26:56,509][starVLA.training.train_starvla][INFO] - Step 1445, Loss: {'train/loss': 0.009147806093096733, 'train/grad_norm_pre_clip': 2.778778076171875, 'timing/data': 0.00029035797342658043, 'timing/model': 1.2310701988171786, 'learning_rate/action_model': 2.1114118492778613e-05, 'learning_rate/action_query_proj': 2.1114118492778613e-05, 'learning_rate/base': 5.278529623194653e-06, 'epoch': 0.62}) +[2026-06-30 05:27:29,182][starVLA.training.train_starvla][INFO] - Step 1446, Loss: {'train/loss': 0.0012135441647842526, 'train/grad_norm_pre_clip': 4.511833667755127, 'timing/data': 0.00020875409245491028, 'timing/model': 1.2309024720452726, 'learning_rate/action_model': 2.105644054587562e-05, 'learning_rate/action_query_proj': 2.105644054587562e-05, 'learning_rate/base': 5.264110136468905e-06, 'epoch': 0.62}) +[2026-06-30 05:28:01,871][starVLA.training.train_starvla][INFO] - Step 1447, Loss: {'train/loss': 0.5762869715690613, 'train/grad_norm_pre_clip': 3.600538492202759, 'timing/data': 0.00019194185733795166, 'timing/model': 1.2325756349600852, 'learning_rate/action_model': 2.0998838949129562e-05, 'learning_rate/action_query_proj': 2.0998838949129562e-05, 'learning_rate/base': 5.249709737282391e-06, 'epoch': 0.62}) +[2026-06-30 05:28:34,563][starVLA.training.train_starvla][INFO] - Step 1448, Loss: {'train/loss': 0.11348523944616318, 'train/grad_norm_pre_clip': 2.876232147216797, 'timing/data': 0.00019305385649204254, 'timing/model': 1.22706134384498, 'learning_rate/action_model': 2.0941313844666687e-05, 'learning_rate/action_query_proj': 2.0941313844666687e-05, 'learning_rate/base': 5.235328461166672e-06, 'epoch': 0.62}) +[2026-06-30 05:29:07,288][starVLA.training.train_starvla][INFO] - Step 1449, Loss: {'train/loss': 0.07456518709659576, 'train/grad_norm_pre_clip': 1.3802651166915894, 'timing/data': 0.00019524781964719296, 'timing/model': 1.234954345971346, 'learning_rate/action_model': 2.0883865374424415e-05, 'learning_rate/action_query_proj': 2.0883865374424415e-05, 'learning_rate/base': 5.220966343606104e-06, 'epoch': 0.62}) +[2026-06-30 05:29:39,946][starVLA.training.train_starvla][INFO] - Step 1450, Loss: {'train/loss': 0.21993254125118256, 'train/grad_norm_pre_clip': 4.594059944152832, 'timing/data': 0.00029237219132483006, 'timing/model': 1.2325673720333725, 'learning_rate/action_model': 2.082649368015119e-05, 'learning_rate/action_query_proj': 2.082649368015119e-05, 'learning_rate/base': 5.206623420037797e-06, 'epoch': 0.62}) +[2026-06-30 05:30:12,651][starVLA.training.train_starvla][INFO] - Step 1451, Loss: {'train/loss': 0.12999746203422546, 'train/grad_norm_pre_clip': 4.749486446380615, 'timing/data': 0.00023800902999937534, 'timing/model': 1.2331258810590953, 'learning_rate/action_model': 2.0769198903405933e-05, 'learning_rate/action_query_proj': 2.0769198903405933e-05, 'learning_rate/base': 5.192299725851483e-06, 'epoch': 0.62}) +[2026-06-30 05:30:45,382][starVLA.training.train_starvla][INFO] - Step 1452, Loss: {'train/loss': 8.200448792194948e-05, 'train/grad_norm_pre_clip': 3.615802526473999, 'timing/data': 0.0001994771882891655, 'timing/model': 1.2345151349436492, 'learning_rate/action_model': 2.0711981185557826e-05, 'learning_rate/action_query_proj': 2.0711981185557826e-05, 'learning_rate/base': 5.1779952963894565e-06, 'epoch': 0.62}) +[2026-06-30 05:31:18,034][starVLA.training.train_starvla][INFO] - Step 1453, Loss: {'train/loss': 0.03673652932047844, 'train/grad_norm_pre_clip': 3.227571964263916, 'timing/data': 0.00027022091671824455, 'timing/model': 1.2313024820759892, 'learning_rate/action_model': 2.0654840667785878e-05, 'learning_rate/action_query_proj': 2.0654840667785878e-05, 'learning_rate/base': 5.1637101669464696e-06, 'epoch': 0.63}) +[2026-06-30 05:31:50,923][starVLA.training.train_starvla][INFO] - Step 1454, Loss: {'train/loss': 0.004816655535250902, 'train/grad_norm_pre_clip': 4.405036926269531, 'timing/data': 0.00043275789357721806, 'timing/model': 1.229880829108879, 'learning_rate/action_model': 2.0597777491078673e-05, 'learning_rate/action_query_proj': 2.0597777491078673e-05, 'learning_rate/base': 5.149444372769668e-06, 'epoch': 0.63}) +[2026-06-30 05:32:23,603][starVLA.training.train_starvla][INFO] - Step 1455, Loss: {'train/loss': 0.00011066596198361367, 'train/grad_norm_pre_clip': 2.641167163848877, 'timing/data': 0.000196842011064291, 'timing/model': 1.2276284359395504, 'learning_rate/action_model': 2.0540791796233898e-05, 'learning_rate/action_query_proj': 2.0540791796233898e-05, 'learning_rate/base': 5.1351979490584746e-06, 'epoch': 0.63}) +[2026-06-30 05:32:56,241][starVLA.training.train_starvla][INFO] - Step 1456, Loss: {'train/loss': 0.016885384917259216, 'train/grad_norm_pre_clip': 4.692306995391846, 'timing/data': 0.0002030930481851101, 'timing/model': 1.2303103830199689, 'learning_rate/action_model': 2.04838837238581e-05, 'learning_rate/action_query_proj': 2.04838837238581e-05, 'learning_rate/base': 5.120970930964525e-06, 'epoch': 0.63}) +[2026-06-30 05:33:28,959][starVLA.training.train_starvla][INFO] - Step 1457, Loss: {'train/loss': 0.0008590350626036525, 'train/grad_norm_pre_clip': 2.871670722961426, 'timing/data': 0.0002902289852499962, 'timing/model': 1.2218210401479155, 'learning_rate/action_model': 2.042705341436627e-05, 'learning_rate/action_query_proj': 2.042705341436627e-05, 'learning_rate/base': 5.106763353591567e-06, 'epoch': 0.63}) +[2026-06-30 05:34:01,599][starVLA.training.train_starvla][INFO] - Step 1458, Loss: {'train/loss': 0.004489287734031677, 'train/grad_norm_pre_clip': 1.923092007637024, 'timing/data': 0.00028824503533542156, 'timing/model': 1.2292876730207354, 'learning_rate/action_model': 2.0370301007981585e-05, 'learning_rate/action_query_proj': 2.0370301007981585e-05, 'learning_rate/base': 5.092575251995396e-06, 'epoch': 0.63}) +[2026-06-30 05:34:34,247][starVLA.training.train_starvla][INFO] - Step 1459, Loss: {'train/loss': 0.0004240618727635592, 'train/grad_norm_pre_clip': 2.7193820476531982, 'timing/data': 0.00020374497398734093, 'timing/model': 1.2180805869866163, 'learning_rate/action_model': 2.031362664473495e-05, 'learning_rate/action_query_proj': 2.031362664473495e-05, 'learning_rate/base': 5.078406661183737e-06, 'epoch': 0.63}) +[2026-06-30 05:35:06,995][starVLA.training.train_starvla][INFO] - Step 1460, Loss: {'train/loss': 0.012117902748286724, 'train/grad_norm_pre_clip': 2.5993149280548096, 'timing/data': 0.0001999279484152794, 'timing/model': 1.2326092720031738, 'learning_rate/action_model': 2.0257030464464714e-05, 'learning_rate/action_query_proj': 2.0257030464464714e-05, 'learning_rate/base': 5.0642576161161786e-06, 'epoch': 0.63}) +[2026-06-30 05:35:39,656][starVLA.training.train_starvla][INFO] - Step 1461, Loss: {'train/loss': 0.06244893744587898, 'train/grad_norm_pre_clip': 3.373232126235962, 'timing/data': 0.00019933609291911125, 'timing/model': 1.231716622132808, 'learning_rate/action_model': 2.0200512606816325e-05, 'learning_rate/action_query_proj': 2.0200512606816325e-05, 'learning_rate/base': 5.050128151704081e-06, 'epoch': 0.63}) +[2026-06-30 05:36:12,339][starVLA.training.train_starvla][INFO] - Step 1462, Loss: {'train/loss': 0.006364875007420778, 'train/grad_norm_pre_clip': 1.4964872598648071, 'timing/data': 0.00020544813014566898, 'timing/model': 1.2300772001035511, 'learning_rate/action_model': 2.0144073211241986e-05, 'learning_rate/action_query_proj': 2.0144073211241986e-05, 'learning_rate/base': 5.0360183028104964e-06, 'epoch': 0.63}) +[2026-06-30 05:36:45,014][starVLA.training.train_starvla][INFO] - Step 1463, Loss: {'train/loss': 2.1338039005058818e-05, 'train/grad_norm_pre_clip': 3.5969812870025635, 'timing/data': 0.00024918094277381897, 'timing/model': 1.2321502319537103, 'learning_rate/action_model': 2.0087712417000304e-05, 'learning_rate/action_query_proj': 2.0087712417000304e-05, 'learning_rate/base': 5.021928104250076e-06, 'epoch': 0.63}) +[2026-06-30 05:37:17,673][starVLA.training.train_starvla][INFO] - Step 1464, Loss: {'train/loss': 0.9682396650314331, 'train/grad_norm_pre_clip': 4.677109718322754, 'timing/data': 0.0002001579850912094, 'timing/model': 1.229806330986321, 'learning_rate/action_model': 2.0031430363155925e-05, 'learning_rate/action_query_proj': 2.0031430363155925e-05, 'learning_rate/base': 5.007857590788981e-06, 'epoch': 0.63}) +[2026-06-30 05:37:50,380][starVLA.training.train_starvla][INFO] - Step 1465, Loss: {'train/loss': 0.019568268209695816, 'train/grad_norm_pre_clip': 1.3824141025543213, 'timing/data': 0.00020390492863953114, 'timing/model': 1.2337920609861612, 'learning_rate/action_model': 1.9975227188579203e-05, 'learning_rate/action_query_proj': 1.9975227188579203e-05, 'learning_rate/base': 4.993806797144801e-06, 'epoch': 0.63}) +[2026-06-30 05:38:23,071][starVLA.training.train_starvla][INFO] - Step 1466, Loss: {'train/loss': 5.8764362620422617e-05, 'train/grad_norm_pre_clip': 1.921190619468689, 'timing/data': 0.00020847399719059467, 'timing/model': 1.233410372864455, 'learning_rate/action_model': 1.9919103031945944e-05, 'learning_rate/action_query_proj': 1.9919103031945944e-05, 'learning_rate/base': 4.979775757986486e-06, 'epoch': 0.63}) +[2026-06-30 05:38:55,766][starVLA.training.train_starvla][INFO] - Step 1467, Loss: {'train/loss': 0.029176900163292885, 'train/grad_norm_pre_clip': 1.4736725091934204, 'timing/data': 0.00019936589524149895, 'timing/model': 1.2347174510359764, 'learning_rate/action_model': 1.986305803173685e-05, 'learning_rate/action_query_proj': 1.986305803173685e-05, 'learning_rate/base': 4.965764507934212e-06, 'epoch': 0.63}) +[2026-06-30 05:39:28,426][starVLA.training.train_starvla][INFO] - Step 1468, Loss: {'train/loss': 0.034068476408720016, 'train/grad_norm_pre_clip': 2.9997429847717285, 'timing/data': 0.0001988660078495741, 'timing/model': 1.235360172111541, 'learning_rate/action_model': 1.980709232623744e-05, 'learning_rate/action_query_proj': 1.980709232623744e-05, 'learning_rate/base': 4.95177308155936e-06, 'epoch': 0.63}) +[2026-06-30 05:40:01,128][starVLA.training.train_starvla][INFO] - Step 1469, Loss: {'train/loss': 0.03835596516728401, 'train/grad_norm_pre_clip': 2.575313091278076, 'timing/data': 0.00028814421966671944, 'timing/model': 1.2285320179071277, 'learning_rate/action_model': 1.975120605353751e-05, 'learning_rate/action_query_proj': 1.975120605353751e-05, 'learning_rate/base': 4.937801513384378e-06, 'epoch': 0.63}) +[2026-06-30 05:40:33,816][starVLA.training.train_starvla][INFO] - Step 1470, Loss: {'train/loss': 0.00352594256401062, 'train/grad_norm_pre_clip': 2.273763656616211, 'timing/data': 0.00019569904543459415, 'timing/model': 1.219118018867448, 'learning_rate/action_model': 1.9695399351530893e-05, 'learning_rate/action_query_proj': 1.9695399351530893e-05, 'learning_rate/base': 4.923849837882723e-06, 'epoch': 0.63}) +[2026-06-30 05:41:06,500][starVLA.training.train_starvla][INFO] - Step 1471, Loss: {'train/loss': 0.0007874781731516123, 'train/grad_norm_pre_clip': 1.4809112548828125, 'timing/data': 0.00024409103207290173, 'timing/model': 1.2305849820841104, 'learning_rate/action_model': 1.963967235791504e-05, 'learning_rate/action_query_proj': 1.963967235791504e-05, 'learning_rate/base': 4.90991808947876e-06, 'epoch': 0.63}) +[2026-06-30 05:41:39,177][starVLA.training.train_starvla][INFO] - Step 1472, Loss: {'train/loss': 8.821443771012127e-06, 'train/grad_norm_pre_clip': 1.749910593032837, 'timing/data': 0.0001962499227374792, 'timing/model': 1.2332530107814819, 'learning_rate/action_model': 1.9584025210190825e-05, 'learning_rate/action_query_proj': 1.9584025210190825e-05, 'learning_rate/base': 4.896006302547706e-06, 'epoch': 0.63}) +[2026-06-30 05:42:11,890][starVLA.training.train_starvla][INFO] - Step 1473, Loss: {'train/loss': 0.20174168050289154, 'train/grad_norm_pre_clip': 1.2758135795593262, 'timing/data': 0.00023914105258882046, 'timing/model': 1.2325309470761567, 'learning_rate/action_model': 1.9528458045662028e-05, 'learning_rate/action_query_proj': 1.9528458045662028e-05, 'learning_rate/base': 4.882114511415507e-06, 'epoch': 0.63}) +[2026-06-30 05:42:44,612][starVLA.training.train_starvla][INFO] - Step 1474, Loss: {'train/loss': 0.00035454981843940914, 'train/grad_norm_pre_clip': 2.1338655948638916, 'timing/data': 0.00024919002316892147, 'timing/model': 1.2297538791317493, 'learning_rate/action_model': 1.9472971001435096e-05, 'learning_rate/action_query_proj': 1.9472971001435096e-05, 'learning_rate/base': 4.868242750358774e-06, 'epoch': 0.63}) +[2026-06-30 05:43:17,299][starVLA.training.train_starvla][INFO] - Step 1475, Loss: {'train/loss': 0.026282677426934242, 'train/grad_norm_pre_clip': 1.0901999473571777, 'timing/data': 0.00020043901167809963, 'timing/model': 1.230975491926074, 'learning_rate/action_model': 1.9417564214418794e-05, 'learning_rate/action_query_proj': 1.9417564214418794e-05, 'learning_rate/base': 4.854391053604699e-06, 'epoch': 0.63}) +[2026-06-30 05:43:49,963][starVLA.training.train_starvla][INFO] - Step 1476, Loss: {'train/loss': 0.05489606782793999, 'train/grad_norm_pre_clip': 2.094989061355591, 'timing/data': 0.00023234891705214977, 'timing/model': 1.2289761931169778, 'learning_rate/action_model': 1.9362237821323875e-05, 'learning_rate/action_query_proj': 1.9362237821323875e-05, 'learning_rate/base': 4.840559455330969e-06, 'epoch': 0.64}) +[2026-06-30 05:44:22,683][starVLA.training.train_starvla][INFO] - Step 1477, Loss: {'train/loss': 0.4410090148448944, 'train/grad_norm_pre_clip': 2.1990115642547607, 'timing/data': 0.00020708097144961357, 'timing/model': 1.2350706809666008, 'learning_rate/action_model': 1.9306991958662717e-05, 'learning_rate/action_query_proj': 1.9306991958662717e-05, 'learning_rate/base': 4.826747989665679e-06, 'epoch': 0.64}) +[2026-06-30 05:44:55,379][starVLA.training.train_starvla][INFO] - Step 1478, Loss: {'train/loss': 0.007618231698870659, 'train/grad_norm_pre_clip': 2.824908971786499, 'timing/data': 0.0003417148254811764, 'timing/model': 1.2320444509387016, 'learning_rate/action_model': 1.9251826762748977e-05, 'learning_rate/action_query_proj': 1.9251826762748977e-05, 'learning_rate/base': 4.812956690687244e-06, 'epoch': 0.64}) +[2026-06-30 05:45:28,044][starVLA.training.train_starvla][INFO] - Step 1479, Loss: {'train/loss': 6.16552570136264e-05, 'train/grad_norm_pre_clip': 2.4720089435577393, 'timing/data': 0.00019039912149310112, 'timing/model': 1.235221290960908, 'learning_rate/action_model': 1.919674236969729e-05, 'learning_rate/action_query_proj': 1.919674236969729e-05, 'learning_rate/base': 4.799185592424323e-06, 'epoch': 0.64}) +[2026-06-30 05:46:00,777][starVLA.training.train_starvla][INFO] - Step 1480, Loss: {'train/loss': 0.018921365961432457, 'train/grad_norm_pre_clip': 0.9205268025398254, 'timing/data': 0.0003003759775310755, 'timing/model': 1.2313336019869894, 'learning_rate/action_model': 1.914173891542294e-05, 'learning_rate/action_query_proj': 1.914173891542294e-05, 'learning_rate/base': 4.785434728855735e-06, 'epoch': 0.64}) +[2026-06-30 05:46:33,487][starVLA.training.train_starvla][INFO] - Step 1481, Loss: {'train/loss': 8.123034058371559e-05, 'train/grad_norm_pre_clip': 2.121676206588745, 'timing/data': 0.0002491602208465338, 'timing/model': 1.232734306016937, 'learning_rate/action_model': 1.9086816535641476e-05, 'learning_rate/action_query_proj': 1.9086816535641476e-05, 'learning_rate/base': 4.771704133910369e-06, 'epoch': 0.64}) +[2026-06-30 05:47:06,158][starVLA.training.train_starvla][INFO] - Step 1482, Loss: {'train/loss': 1.61526222655084e-05, 'train/grad_norm_pre_clip': 0.9999797940254211, 'timing/data': 0.0003432268276810646, 'timing/model': 1.2313521418254822, 'learning_rate/action_model': 1.903197536586841e-05, 'learning_rate/action_query_proj': 1.903197536586841e-05, 'learning_rate/base': 4.757993841467102e-06, 'epoch': 0.64}) +[2026-06-30 05:47:38,868][starVLA.training.train_starvla][INFO] - Step 1483, Loss: {'train/loss': 5.0481397920520976e-05, 'train/grad_norm_pre_clip': 1.7285147905349731, 'timing/data': 0.00022591720335185528, 'timing/model': 1.2326531040016562, 'learning_rate/action_model': 1.897721554141886e-05, 'learning_rate/action_query_proj': 1.897721554141886e-05, 'learning_rate/base': 4.744303885354715e-06, 'epoch': 0.64}) +[2026-06-30 05:48:11,554][starVLA.training.train_starvla][INFO] - Step 1484, Loss: {'train/loss': 0.21176321804523468, 'train/grad_norm_pre_clip': 2.047100067138672, 'timing/data': 0.00026087299920618534, 'timing/model': 1.2275748618412763, 'learning_rate/action_model': 1.892253719740728e-05, 'learning_rate/action_query_proj': 1.892253719740728e-05, 'learning_rate/base': 4.73063429935182e-06, 'epoch': 0.64}) +[2026-06-30 05:48:44,206][starVLA.training.train_starvla][INFO] - Step 1485, Loss: {'train/loss': 0.09729600697755814, 'train/grad_norm_pre_clip': 1.4403821229934692, 'timing/data': 0.0002011291217058897, 'timing/model': 1.229257077910006, 'learning_rate/action_model': 1.8867940468747015e-05, 'learning_rate/action_query_proj': 1.8867940468747015e-05, 'learning_rate/base': 4.716985117186754e-06, 'epoch': 0.64}) +[2026-06-30 05:49:16,878][starVLA.training.train_starvla][INFO] - Step 1486, Loss: {'train/loss': 0.06422917544841766, 'train/grad_norm_pre_clip': 2.7467894554138184, 'timing/data': 0.00020803301595151424, 'timing/model': 1.2360816600266844, 'learning_rate/action_model': 1.8813425490150093e-05, 'learning_rate/action_query_proj': 1.8813425490150093e-05, 'learning_rate/base': 4.703356372537523e-06, 'epoch': 0.64}) +[2026-06-30 05:49:49,579][starVLA.training.train_starvla][INFO] - Step 1487, Loss: {'train/loss': 0.09491223096847534, 'train/grad_norm_pre_clip': 2.758479356765747, 'timing/data': 0.0004705179017037153, 'timing/model': 1.2322221740614623, 'learning_rate/action_model': 1.8758992396126795e-05, 'learning_rate/action_query_proj': 1.8758992396126795e-05, 'learning_rate/base': 4.689748099031699e-06, 'epoch': 0.64}) +[2026-06-30 05:50:22,306][starVLA.training.train_starvla][INFO] - Step 1488, Loss: {'train/loss': 0.006681123282760382, 'train/grad_norm_pre_clip': 2.7180356979370117, 'timing/data': 0.00019746203906834126, 'timing/model': 1.2289562509395182, 'learning_rate/action_model': 1.8704641320985367e-05, 'learning_rate/action_query_proj': 1.8704641320985367e-05, 'learning_rate/base': 4.676160330246342e-06, 'epoch': 0.64}) +[2026-06-30 05:50:54,951][starVLA.training.train_starvla][INFO] - Step 1489, Loss: {'train/loss': 0.0001865097729023546, 'train/grad_norm_pre_clip': 1.2332563400268555, 'timing/data': 0.00020611914806067944, 'timing/model': 1.2326954170130193, 'learning_rate/action_model': 1.8650372398831674e-05, 'learning_rate/action_query_proj': 1.8650372398831674e-05, 'learning_rate/base': 4.6625930997079185e-06, 'epoch': 0.64}) +[2026-06-30 05:51:27,644][starVLA.training.train_starvla][INFO] - Step 1490, Loss: {'train/loss': 0.01498178206384182, 'train/grad_norm_pre_clip': 2.230273485183716, 'timing/data': 0.00020217103883624077, 'timing/model': 1.2345789629034698, 'learning_rate/action_model': 1.859618576356891e-05, 'learning_rate/action_query_proj': 1.859618576356891e-05, 'learning_rate/base': 4.649046440892228e-06, 'epoch': 0.64}) +[2026-06-30 05:52:00,316][starVLA.training.train_starvla][INFO] - Step 1491, Loss: {'train/loss': 0.0027276810724288225, 'train/grad_norm_pre_clip': 4.0071539878845215, 'timing/data': 0.00031387200579047203, 'timing/model': 1.2288468219339848, 'learning_rate/action_model': 1.85420815488972e-05, 'learning_rate/action_query_proj': 1.85420815488972e-05, 'learning_rate/base': 4.6355203872243e-06, 'epoch': 0.64}) +[2026-06-30 05:52:33,006][starVLA.training.train_starvla][INFO] - Step 1492, Loss: {'train/loss': 0.0006564230425283313, 'train/grad_norm_pre_clip': 2.084153413772583, 'timing/data': 0.0003610709682106972, 'timing/model': 1.2299452109728009, 'learning_rate/action_model': 1.848805988831332e-05, 'learning_rate/action_query_proj': 1.848805988831332e-05, 'learning_rate/base': 4.62201497207833e-06, 'epoch': 0.64}) +[2026-06-30 05:53:05,649][starVLA.training.train_starvla][INFO] - Step 1493, Loss: {'train/loss': 0.0002483731950633228, 'train/grad_norm_pre_clip': 2.178283929824829, 'timing/data': 0.00022313115186989307, 'timing/model': 1.2315401071682572, 'learning_rate/action_model': 1.843412091511032e-05, 'learning_rate/action_query_proj': 1.843412091511032e-05, 'learning_rate/base': 4.60853022877758e-06, 'epoch': 0.64}) +[2026-06-30 05:53:38,348][starVLA.training.train_starvla][INFO] - Step 1494, Loss: {'train/loss': 0.000287645059870556, 'train/grad_norm_pre_clip': 3.151524782180786, 'timing/data': 0.00019478797912597656, 'timing/model': 1.2293415579479188, 'learning_rate/action_model': 1.83802647623773e-05, 'learning_rate/action_query_proj': 1.83802647623773e-05, 'learning_rate/base': 4.595066190594325e-06, 'epoch': 0.64}) +[2026-06-30 05:54:10,992][starVLA.training.train_starvla][INFO] - Step 1495, Loss: {'train/loss': 0.09252568334341049, 'train/grad_norm_pre_clip': 1.7529891729354858, 'timing/data': 0.0002377689816057682, 'timing/model': 1.2343131680972874, 'learning_rate/action_model': 1.8326491562998927e-05, 'learning_rate/action_query_proj': 1.8326491562998927e-05, 'learning_rate/base': 4.581622890749732e-06, 'epoch': 0.64}) +[2026-06-30 05:54:43,723][starVLA.training.train_starvla][INFO] - Step 1496, Loss: {'train/loss': 5.063042772235349e-05, 'train/grad_norm_pre_clip': 2.564554452896118, 'timing/data': 0.00038378406316041946, 'timing/model': 1.228136254940182, 'learning_rate/action_model': 1.827280144965524e-05, 'learning_rate/action_query_proj': 1.827280144965524e-05, 'learning_rate/base': 4.56820036241381e-06, 'epoch': 0.64}) +[2026-06-30 05:55:16,439][starVLA.training.train_starvla][INFO] - Step 1497, Loss: {'train/loss': 0.032105185091495514, 'train/grad_norm_pre_clip': 2.2760794162750244, 'timing/data': 0.000199194997549057, 'timing/model': 1.2343780628871173, 'learning_rate/action_model': 1.8219194554821245e-05, 'learning_rate/action_query_proj': 1.8219194554821245e-05, 'learning_rate/base': 4.554798638705311e-06, 'epoch': 0.64}) +[2026-06-30 05:55:49,098][starVLA.training.train_starvla][INFO] - Step 1498, Loss: {'train/loss': 0.024744756519794464, 'train/grad_norm_pre_clip': 2.3428573608398438, 'timing/data': 0.0002472170162945986, 'timing/model': 1.2330550809856504, 'learning_rate/action_model': 1.8165671010766648e-05, 'learning_rate/action_query_proj': 1.8165671010766648e-05, 'learning_rate/base': 4.541417752691662e-06, 'epoch': 0.64}) +[2026-06-30 05:56:21,716][starVLA.training.train_starvla][INFO] - Step 1499, Loss: {'train/loss': 6.139252491266234e-06, 'train/grad_norm_pre_clip': 2.0776114463806152, 'timing/data': 0.00020076916553080082, 'timing/model': 1.2334979008883238, 'learning_rate/action_model': 1.8112230949555458e-05, 'learning_rate/action_query_proj': 1.8112230949555458e-05, 'learning_rate/base': 4.5280577373888646e-06, 'epoch': 0.64}) +[2026-06-30 05:56:54,431][starVLA.training.train_starvla][INFO] - Starting held-out action loss eval at step 1500 +[2026-06-30 05:57:31,689][starVLA.training.train_starvla][INFO] - Finished held-out action loss eval at step 1500 +[2026-06-30 05:57:31,689][starVLA.training.train_starvla][INFO] - Starting held-out action classification eval at step 1500 +[2026-06-30 05:57:31,690][starVLA.training.train_starvla][INFO] - Finished held-out action classification eval at step 1500 +[2026-06-30 06:46:06,057][starVLA.training.rl_games.checkpoint_sync][INFO] - HF checkpoint sync disabled; skipping eval result upload for step 1500 +[2026-06-30 06:46:06,057][starVLA.training.train_starvla][INFO] - Step 1500, Loss: {'train/loss': 0.00016766709450166672, 'train/grad_norm_pre_clip': 1.7167936563491821, 'eval/action_loss/seconds': 37.25502027501352, 'eval/action_loss/samples': 200.0, 'eval/loss': 0.021576959639787674, 'eval/latency_1/loss': 0.021576959639787674, 'eval/flappy/loss': 0.021576959639787674, 'eval/flappy/latency_1/loss': 0.021576959639787674, 'rl_games_eval/mid_train/total_episodes': 5.0, 'rl_games_eval/mid_train/mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/mean_length': 3600.0, 'rl_games_eval/mid_train/std_reward': 0.0, 'rl_games_eval/mid_train/std_length': 0.0, 'rl_games_eval/mid_train/macro_mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/macro_mean_length': 3600.0, 'rl_games_eval/mid_train/task_count': 1.0, 'rl_games_eval/mid_train/flappy__latency_1/mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/flappy__latency_1/mean_length': 3600.0, 'rl_games_eval/mid_train/flappy__latency_1/std_reward': 0.0, 'rl_games_eval/mid_train/flappy__latency_1/std_length': 0.0, 'rl_games_eval/mid_train/flappy__latency_1/num_episodes': 5.0, 'checkpoint/current_eval_score': 444.60000000002566, 'checkpoint/best_score': -inf, 'checkpoint/best_step': 0.0, 'checkpoint/is_new_best': 0.0, 'timing/data': 0.000199456000700593, 'timing/model': 1.2301137920003384, 'learning_rate/action_model': 1.8058874503045718e-05, 'learning_rate/action_query_proj': 1.8058874503045718e-05, 'learning_rate/base': 4.5147186257614295e-06, 'epoch': 0.65}) +[2026-06-30 06:46:38,776][starVLA.training.train_starvla][INFO] - Step 1501, Loss: {'train/loss': 3.4599059290485457e-05, 'train/grad_norm_pre_clip': 2.3464581966400146, 'timing/data': 0.00019614002667367458, 'timing/model': 1.2328240780625492, 'learning_rate/action_model': 1.800560180288916e-05, 'learning_rate/action_query_proj': 1.800560180288916e-05, 'learning_rate/base': 4.50140045072229e-06, 'epoch': 0.65}) +[2026-06-30 06:47:11,439][starVLA.training.train_starvla][INFO] - Step 1502, Loss: {'train/loss': 0.041541699320077896, 'train/grad_norm_pre_clip': 1.8775686025619507, 'timing/data': 0.0001965109258890152, 'timing/model': 1.2332820519804955, 'learning_rate/action_model': 1.7952412980530863e-05, 'learning_rate/action_query_proj': 1.7952412980530863e-05, 'learning_rate/base': 4.488103245132716e-06, 'epoch': 0.65}) +[2026-06-30 06:47:44,113][starVLA.training.train_starvla][INFO] - Step 1503, Loss: {'train/loss': 2.0444125766516663e-05, 'train/grad_norm_pre_clip': 3.274683952331543, 'timing/data': 0.0003285198472440243, 'timing/model': 1.23230638098903, 'learning_rate/action_model': 1.7899308167208968e-05, 'learning_rate/action_query_proj': 1.7899308167208968e-05, 'learning_rate/base': 4.474827041802242e-06, 'epoch': 0.65}) +[2026-06-30 06:48:16,967][starVLA.training.train_starvla][INFO] - Step 1504, Loss: {'train/loss': 0.04422824829816818, 'train/grad_norm_pre_clip': 3.4619874954223633, 'timing/data': 0.0004359020385891199, 'timing/model': 1.2322754298802465, 'learning_rate/action_model': 1.7846287493954307e-05, 'learning_rate/action_query_proj': 1.7846287493954307e-05, 'learning_rate/base': 4.461571873488577e-06, 'epoch': 0.65}) +[2026-06-30 06:48:49,620][starVLA.training.train_starvla][INFO] - Step 1505, Loss: {'train/loss': 0.03098892793059349, 'train/grad_norm_pre_clip': 2.7907941341400146, 'timing/data': 0.00024924008175730705, 'timing/model': 1.231531301047653, 'learning_rate/action_model': 1.7793351091590144e-05, 'learning_rate/action_query_proj': 1.7793351091590144e-05, 'learning_rate/base': 4.448337772897536e-06, 'epoch': 0.65}) +[2026-06-30 06:49:22,284][starVLA.training.train_starvla][INFO] - Step 1506, Loss: {'train/loss': 0.0036307526752352715, 'train/grad_norm_pre_clip': 1.524505615234375, 'timing/data': 0.00021958397701382637, 'timing/model': 1.2327139030676335, 'learning_rate/action_model': 1.7740499090731782e-05, 'learning_rate/action_query_proj': 1.7740499090731782e-05, 'learning_rate/base': 4.4351247726829455e-06, 'epoch': 0.65}) +[2026-06-30 06:49:54,953][starVLA.training.train_starvla][INFO] - Step 1507, Loss: {'train/loss': 0.013817762956023216, 'train/grad_norm_pre_clip': 2.873039960861206, 'timing/data': 0.0003262248355895281, 'timing/model': 1.228027073899284, 'learning_rate/action_model': 1.768773162178628e-05, 'learning_rate/action_query_proj': 1.768773162178628e-05, 'learning_rate/base': 4.42193290544657e-06, 'epoch': 0.65}) +[2026-06-30 06:50:27,601][starVLA.training.train_starvla][INFO] - Step 1508, Loss: {'train/loss': 0.0017556448001414537, 'train/grad_norm_pre_clip': 2.836482048034668, 'timing/data': 0.000414832029491663, 'timing/model': 1.228997382801026, 'learning_rate/action_model': 1.7635048814952097e-05, 'learning_rate/action_query_proj': 1.7635048814952097e-05, 'learning_rate/base': 4.408762203738024e-06, 'epoch': 0.65}) +[2026-06-30 06:51:00,234][starVLA.training.train_starvla][INFO] - Step 1509, Loss: {'train/loss': 3.921759343938902e-05, 'train/grad_norm_pre_clip': 2.04259991645813, 'timing/data': 0.0001951879821717739, 'timing/model': 1.2283803049940616, 'learning_rate/action_model': 1.758245080021887e-05, 'learning_rate/action_query_proj': 1.758245080021887e-05, 'learning_rate/base': 4.395612700054717e-06, 'epoch': 0.65}) +[2026-06-30 06:51:32,867][starVLA.training.train_starvla][INFO] - Step 1510, Loss: {'train/loss': 0.00028234414639882743, 'train/grad_norm_pre_clip': 1.463430404663086, 'timing/data': 0.00020718108862638474, 'timing/model': 1.2288527400232852, 'learning_rate/action_model': 1.7529937707366946e-05, 'learning_rate/action_query_proj': 1.7529937707366946e-05, 'learning_rate/base': 4.3824844268417365e-06, 'epoch': 0.65}) +[2026-06-30 06:52:05,528][starVLA.training.train_starvla][INFO] - Step 1511, Loss: {'train/loss': 0.0035845462698489428, 'train/grad_norm_pre_clip': 3.1512057781219482, 'timing/data': 0.00020387396216392517, 'timing/model': 1.2318264220375568, 'learning_rate/action_model': 1.747750966596715e-05, 'learning_rate/action_query_proj': 1.747750966596715e-05, 'learning_rate/base': 4.369377416491788e-06, 'epoch': 0.65}) +[2026-06-30 06:52:38,162][starVLA.training.train_starvla][INFO] - Step 1512, Loss: {'train/loss': 3.4420343581587076e-05, 'train/grad_norm_pre_clip': 1.9469362497329712, 'timing/data': 0.00029207090847194195, 'timing/model': 1.2287625258322805, 'learning_rate/action_model': 1.7425166805380465e-05, 'learning_rate/action_query_proj': 1.7425166805380465e-05, 'learning_rate/base': 4.356291701345116e-06, 'epoch': 0.65}) +[2026-06-30 06:53:10,838][starVLA.training.train_starvla][INFO] - Step 1513, Loss: {'train/loss': 0.1584172546863556, 'train/grad_norm_pre_clip': 3.1609177589416504, 'timing/data': 0.0002889449242502451, 'timing/model': 1.231083112070337, 'learning_rate/action_model': 1.7372909254757724e-05, 'learning_rate/action_query_proj': 1.7372909254757724e-05, 'learning_rate/base': 4.343227313689431e-06, 'epoch': 0.65}) +[2026-06-30 06:53:43,508][starVLA.training.train_starvla][INFO] - Step 1514, Loss: {'train/loss': 0.001524797291494906, 'train/grad_norm_pre_clip': 3.846738338470459, 'timing/data': 0.0004505389370024204, 'timing/model': 1.2288597780279815, 'learning_rate/action_model': 1.7320737143039212e-05, 'learning_rate/action_query_proj': 1.7320737143039212e-05, 'learning_rate/base': 4.330184285759803e-06, 'epoch': 0.65}) +[2026-06-30 06:54:16,166][starVLA.training.train_starvla][INFO] - Step 1515, Loss: {'train/loss': 0.13160601258277893, 'train/grad_norm_pre_clip': 2.3656511306762695, 'timing/data': 0.00020257220603525639, 'timing/model': 1.2279520940501243, 'learning_rate/action_model': 1.726865059895444e-05, 'learning_rate/action_query_proj': 1.726865059895444e-05, 'learning_rate/base': 4.31716264973861e-06, 'epoch': 0.65}) +[2026-06-30 06:54:48,826][starVLA.training.train_starvla][INFO] - Step 1516, Loss: {'train/loss': 0.0030330875888466835, 'train/grad_norm_pre_clip': 2.962442636489868, 'timing/data': 0.0002626448404043913, 'timing/model': 1.2325746698770672, 'learning_rate/action_model': 1.7216649751021757e-05, 'learning_rate/action_query_proj': 1.7216649751021757e-05, 'learning_rate/base': 4.304162437755439e-06, 'epoch': 0.65}) +[2026-06-30 06:55:21,472][starVLA.training.train_starvla][INFO] - Step 1517, Loss: {'train/loss': 0.3879321813583374, 'train/grad_norm_pre_clip': 3.7226555347442627, 'timing/data': 0.00019940617494285107, 'timing/model': 1.228397788014263, 'learning_rate/action_model': 1.7164734727548133e-05, 'learning_rate/action_query_proj': 1.7164734727548133e-05, 'learning_rate/base': 4.291183681887033e-06, 'epoch': 0.65}) +[2026-06-30 06:55:54,148][starVLA.training.train_starvla][INFO] - Step 1518, Loss: {'train/loss': 0.130722314119339, 'train/grad_norm_pre_clip': 3.5209789276123047, 'timing/data': 0.00021011615172028542, 'timing/model': 1.2335131601430476, 'learning_rate/action_model': 1.711290565662867e-05, 'learning_rate/action_query_proj': 1.711290565662867e-05, 'learning_rate/base': 4.278226414157167e-06, 'epoch': 0.65}) +[2026-06-30 06:56:26,804][starVLA.training.train_starvla][INFO] - Step 1519, Loss: {'train/loss': 0.0033523663878440857, 'train/grad_norm_pre_clip': 1.4140573740005493, 'timing/data': 0.0003449399955570698, 'timing/model': 1.23062978615053, 'learning_rate/action_model': 1.7061162666146484e-05, 'learning_rate/action_query_proj': 1.7061162666146484e-05, 'learning_rate/base': 4.265290666536621e-06, 'epoch': 0.65}) +[2026-06-30 06:56:59,447][starVLA.training.train_starvla][INFO] - Step 1520, Loss: {'train/loss': 0.002263080095872283, 'train/grad_norm_pre_clip': 4.027675151824951, 'timing/data': 0.00026019103825092316, 'timing/model': 1.2356554679572582, 'learning_rate/action_model': 1.7009505883772253e-05, 'learning_rate/action_query_proj': 1.7009505883772253e-05, 'learning_rate/base': 4.252376470943063e-06, 'epoch': 0.65}) +[2026-06-30 06:57:32,084][starVLA.training.train_starvla][INFO] - Step 1521, Loss: {'train/loss': 0.010035118088126183, 'train/grad_norm_pre_clip': 4.641293048858643, 'timing/data': 0.0002763010561466217, 'timing/model': 1.23167949100025, 'learning_rate/action_model': 1.6957935436963957e-05, 'learning_rate/action_query_proj': 1.6957935436963957e-05, 'learning_rate/base': 4.239483859240989e-06, 'epoch': 0.65}) +[2026-06-30 06:58:04,759][starVLA.training.train_starvla][INFO] - Step 1522, Loss: {'train/loss': 1.2934113328810781e-05, 'train/grad_norm_pre_clip': 3.5219154357910156, 'timing/data': 0.0003319159150123596, 'timing/model': 1.2346468889154494, 'learning_rate/action_model': 1.690645145296652e-05, 'learning_rate/action_query_proj': 1.690645145296652e-05, 'learning_rate/base': 4.22661286324163e-06, 'epoch': 0.65}) +[2026-06-30 06:58:37,462][starVLA.training.train_starvla][INFO] - Step 1523, Loss: {'train/loss': 0.008456328883767128, 'train/grad_norm_pre_clip': 2.2859601974487305, 'timing/data': 0.00020095892250537872, 'timing/model': 1.2337717988993973, 'learning_rate/action_model': 1.6855054058811596e-05, 'learning_rate/action_query_proj': 1.6855054058811596e-05, 'learning_rate/base': 4.213763514702899e-06, 'epoch': 0.66}) +[2026-06-30 06:59:10,124][starVLA.training.train_starvla][INFO] - Step 1524, Loss: {'train/loss': 0.017522791400551796, 'train/grad_norm_pre_clip': 2.4430038928985596, 'timing/data': 0.00045334501191973686, 'timing/model': 1.2287100728135556, 'learning_rate/action_model': 1.6803743381317127e-05, 'learning_rate/action_query_proj': 1.6803743381317127e-05, 'learning_rate/base': 4.200935845329282e-06, 'epoch': 0.66}) +[2026-06-30 06:59:42,774][starVLA.training.train_starvla][INFO] - Step 1525, Loss: {'train/loss': 0.18199558556079865, 'train/grad_norm_pre_clip': 1.5383678674697876, 'timing/data': 0.0001994660124182701, 'timing/model': 1.2309333360753953, 'learning_rate/action_model': 1.6752519547087103e-05, 'learning_rate/action_query_proj': 1.6752519547087103e-05, 'learning_rate/base': 4.188129886771776e-06, 'epoch': 0.66}) +[2026-06-30 07:00:15,457][starVLA.training.train_starvla][INFO] - Step 1526, Loss: {'train/loss': 0.0319780632853508, 'train/grad_norm_pre_clip': 2.800233840942383, 'timing/data': 0.00022523500956594944, 'timing/model': 1.229046086082235, 'learning_rate/action_model': 1.6701382682511234e-05, 'learning_rate/action_query_proj': 1.6701382682511234e-05, 'learning_rate/base': 4.1753456706278085e-06, 'epoch': 0.66}) +[2026-06-30 07:00:48,110][starVLA.training.train_starvla][INFO] - Step 1527, Loss: {'train/loss': 0.07220534980297089, 'train/grad_norm_pre_clip': 1.815064787864685, 'timing/data': 0.0002001868560910225, 'timing/model': 1.2319216078612953, 'learning_rate/action_model': 1.6650332913764683e-05, 'learning_rate/action_query_proj': 1.6650332913764683e-05, 'learning_rate/base': 4.162583228441171e-06, 'epoch': 0.66}) +[2026-06-30 07:01:20,734][starVLA.training.train_starvla][INFO] - Step 1528, Loss: {'train/loss': 0.04254715517163277, 'train/grad_norm_pre_clip': 3.7604143619537354, 'timing/data': 0.0003054458647966385, 'timing/model': 1.22964774700813, 'learning_rate/action_model': 1.6599370366807655e-05, 'learning_rate/action_query_proj': 1.6599370366807655e-05, 'learning_rate/base': 4.149842591701914e-06, 'epoch': 0.66}) +[2026-06-30 07:01:53,360][starVLA.training.train_starvla][INFO] - Step 1529, Loss: {'train/loss': 0.01337283756583929, 'train/grad_norm_pre_clip': 1.7582173347473145, 'timing/data': 0.00020274287089705467, 'timing/model': 1.2323646279983222, 'learning_rate/action_model': 1.654849516738518e-05, 'learning_rate/action_query_proj': 1.654849516738518e-05, 'learning_rate/base': 4.137123791846295e-06, 'epoch': 0.66}) +[2026-06-30 07:02:26,010][starVLA.training.train_starvla][INFO] - Step 1530, Loss: {'train/loss': 0.2245161533355713, 'train/grad_norm_pre_clip': 2.1052980422973633, 'timing/data': 0.00020483601838350296, 'timing/model': 1.2275120329577476, 'learning_rate/action_model': 1.649770744102673e-05, 'learning_rate/action_query_proj': 1.649770744102673e-05, 'learning_rate/base': 4.124426860256683e-06, 'epoch': 0.66}) +[2026-06-30 07:02:58,664][starVLA.training.train_starvla][INFO] - Step 1531, Loss: {'train/loss': 1.940103538800031e-05, 'train/grad_norm_pre_clip': 3.38936710357666, 'timing/data': 0.00045969802886247635, 'timing/model': 1.2291858061216772, 'learning_rate/action_model': 1.6447007313046006e-05, 'learning_rate/action_query_proj': 1.6447007313046006e-05, 'learning_rate/base': 4.1117518282615014e-06, 'epoch': 0.66}) +[2026-06-30 07:03:31,343][starVLA.training.train_starvla][INFO] - Step 1532, Loss: {'train/loss': 4.580372842610814e-05, 'train/grad_norm_pre_clip': 2.232160806655884, 'timing/data': 0.00025687413290143013, 'timing/model': 1.2307053380645812, 'learning_rate/action_model': 1.6396394908540524e-05, 'learning_rate/action_query_proj': 1.6396394908540524e-05, 'learning_rate/base': 4.099098727135131e-06, 'epoch': 0.66}) +[2026-06-30 07:04:03,986][starVLA.training.train_starvla][INFO] - Step 1533, Loss: {'train/loss': 0.000168562401086092, 'train/grad_norm_pre_clip': 2.133671522140503, 'timing/data': 0.00020275195129215717, 'timing/model': 1.2285025289747864, 'learning_rate/action_model': 1.6345870352391348e-05, 'learning_rate/action_query_proj': 1.6345870352391348e-05, 'learning_rate/base': 4.086467588097837e-06, 'epoch': 0.66}) +[2026-06-30 07:04:36,631][starVLA.training.train_starvla][INFO] - Step 1534, Loss: {'train/loss': 0.0018442575819790363, 'train/grad_norm_pre_clip': 4.419704437255859, 'timing/data': 0.00020403414964675903, 'timing/model': 1.2330437791533768, 'learning_rate/action_model': 1.6295433769262794e-05, 'learning_rate/action_query_proj': 1.6295433769262794e-05, 'learning_rate/base': 4.073858442315698e-06, 'epoch': 0.66}) +[2026-06-30 07:05:09,269][starVLA.training.train_starvla][INFO] - Step 1535, Loss: {'train/loss': 0.3621138334274292, 'train/grad_norm_pre_clip': 2.8964362144470215, 'timing/data': 0.0003451609518378973, 'timing/model': 1.2327639069408178, 'learning_rate/action_model': 1.6245085283602155e-05, 'learning_rate/action_query_proj': 1.6245085283602155e-05, 'learning_rate/base': 4.061271320900539e-06, 'epoch': 0.66}) +[2026-06-30 07:05:41,910][starVLA.training.train_starvla][INFO] - Step 1536, Loss: {'train/loss': 6.812301580794156e-05, 'train/grad_norm_pre_clip': 1.7817245721817017, 'timing/data': 0.0001978529617190361, 'timing/model': 1.2334575799759477, 'learning_rate/action_model': 1.619482501963926e-05, 'learning_rate/action_query_proj': 1.619482501963926e-05, 'learning_rate/base': 4.048706254909815e-06, 'epoch': 0.66}) +[2026-06-30 07:06:14,601][starVLA.training.train_starvla][INFO] - Step 1537, Loss: {'train/loss': 0.00035230052890256047, 'train/grad_norm_pre_clip': 2.9275457859039307, 'timing/data': 0.0001944471150636673, 'timing/model': 1.229430329054594, 'learning_rate/action_model': 1.6144653101386354e-05, 'learning_rate/action_query_proj': 1.6144653101386354e-05, 'learning_rate/base': 4.036163275346588e-06, 'epoch': 0.66}) +[2026-06-30 07:06:47,241][starVLA.training.train_starvla][INFO] - Step 1538, Loss: {'train/loss': 0.047133829444646835, 'train/grad_norm_pre_clip': 3.8621602058410645, 'timing/data': 0.0001923020463436842, 'timing/model': 1.2309526838362217, 'learning_rate/action_model': 1.609456965263765e-05, 'learning_rate/action_query_proj': 1.609456965263765e-05, 'learning_rate/base': 4.023642413159413e-06, 'epoch': 0.66}) +[2026-06-30 07:07:19,859][starVLA.training.train_starvla][INFO] - Step 1539, Loss: {'train/loss': 0.2658856511116028, 'train/grad_norm_pre_clip': 2.1140964031219482, 'timing/data': 0.00020332401618361473, 'timing/model': 1.2275145628955215, 'learning_rate/action_model': 1.6044574796969066e-05, 'learning_rate/action_query_proj': 1.6044574796969066e-05, 'learning_rate/base': 4.0111436992422665e-06, 'epoch': 0.66}) +[2026-06-30 07:07:52,515][starVLA.training.train_starvla][INFO] - Step 1540, Loss: {'train/loss': 3.829380148090422e-05, 'train/grad_norm_pre_clip': 4.157118320465088, 'timing/data': 0.0002997259143739939, 'timing/model': 1.2287701950408518, 'learning_rate/action_model': 1.5994668657737936e-05, 'learning_rate/action_query_proj': 1.5994668657737936e-05, 'learning_rate/base': 3.998667164434484e-06, 'epoch': 0.66}) +[2026-06-30 07:08:25,166][starVLA.training.train_starvla][INFO] - Step 1541, Loss: {'train/loss': 8.900782268028706e-05, 'train/grad_norm_pre_clip': 2.269946575164795, 'timing/data': 0.00030755018815398216, 'timing/model': 1.232849532039836, 'learning_rate/action_model': 1.594485135808273e-05, 'learning_rate/action_query_proj': 1.594485135808273e-05, 'learning_rate/base': 3.986212839520683e-06, 'epoch': 0.66}) +[2026-06-30 07:08:57,853][starVLA.training.train_starvla][INFO] - Step 1542, Loss: {'train/loss': 0.003843693993985653, 'train/grad_norm_pre_clip': 1.2319904565811157, 'timing/data': 0.00035501900129020214, 'timing/model': 1.2282160960603505, 'learning_rate/action_model': 1.589512302092266e-05, 'learning_rate/action_query_proj': 1.589512302092266e-05, 'learning_rate/base': 3.973780755230665e-06, 'epoch': 0.66}) +[2026-06-30 07:09:30,562][starVLA.training.train_starvla][INFO] - Step 1543, Loss: {'train/loss': 0.3416266143321991, 'train/grad_norm_pre_clip': 2.0751774311065674, 'timing/data': 0.00020068790763616562, 'timing/model': 1.2264786618761718, 'learning_rate/action_model': 1.584548376895745e-05, 'learning_rate/action_query_proj': 1.584548376895745e-05, 'learning_rate/base': 3.961370942239363e-06, 'epoch': 0.66}) +[2026-06-30 07:10:03,173][starVLA.training.train_starvla][INFO] - Step 1544, Loss: {'train/loss': 5.2626342949224636e-05, 'train/grad_norm_pre_clip': 1.533328890800476, 'timing/data': 0.00022295094095170498, 'timing/model': 1.2310338669922203, 'learning_rate/action_model': 1.5795933724667013e-05, 'learning_rate/action_query_proj': 1.5795933724667013e-05, 'learning_rate/base': 3.948983431166753e-06, 'epoch': 0.66}) +[2026-06-30 07:10:35,808][starVLA.training.train_starvla][INFO] - Step 1545, Loss: {'train/loss': 0.05169752240180969, 'train/grad_norm_pre_clip': 3.3985848426818848, 'timing/data': 0.00025888788513839245, 'timing/model': 1.229771618032828, 'learning_rate/action_model': 1.574647301031119e-05, 'learning_rate/action_query_proj': 1.574647301031119e-05, 'learning_rate/base': 3.936618252577798e-06, 'epoch': 0.66}) +[2026-06-30 07:11:08,410][starVLA.training.train_starvla][INFO] - Step 1546, Loss: {'train/loss': 0.0024487231858074665, 'train/grad_norm_pre_clip': 2.0491104125976562, 'timing/data': 0.00019916589371860027, 'timing/model': 1.2274719539564103, 'learning_rate/action_model': 1.5697101747929356e-05, 'learning_rate/action_query_proj': 1.5697101747929356e-05, 'learning_rate/base': 3.924275436982339e-06, 'epoch': 0.67}) +[2026-06-30 07:11:41,047][starVLA.training.train_starvla][INFO] - Step 1547, Loss: {'train/loss': 0.003580590710043907, 'train/grad_norm_pre_clip': 2.1641416549682617, 'timing/data': 0.0003065878991037607, 'timing/model': 1.2364470239263028, 'learning_rate/action_model': 1.564782005934019e-05, 'learning_rate/action_query_proj': 1.564782005934019e-05, 'learning_rate/base': 3.911955014835048e-06, 'epoch': 0.67}) +[2026-06-30 07:12:13,729][starVLA.training.train_starvla][INFO] - Step 1548, Loss: {'train/loss': 0.007164506241679192, 'train/grad_norm_pre_clip': 3.1660234928131104, 'timing/data': 0.00020403484813869, 'timing/model': 1.2309389568399638, 'learning_rate/action_model': 1.559862806614135e-05, 'learning_rate/action_query_proj': 1.559862806614135e-05, 'learning_rate/base': 3.899657016535338e-06, 'epoch': 0.67}) +[2026-06-30 07:12:46,366][starVLA.training.train_starvla][INFO] - Step 1549, Loss: {'train/loss': 0.0008385921246372163, 'train/grad_norm_pre_clip': 3.2776429653167725, 'timing/data': 0.00024232803843915462, 'timing/model': 1.2292906711809337, 'learning_rate/action_model': 1.5549525889709214e-05, 'learning_rate/action_query_proj': 1.5549525889709214e-05, 'learning_rate/base': 3.8873814724273035e-06, 'epoch': 0.67}) +[2026-06-30 07:13:19,009][starVLA.training.train_starvla][INFO] - Step 1550, Loss: {'train/loss': 0.13728077709674835, 'train/grad_norm_pre_clip': 1.9359469413757324, 'timing/data': 0.00018916791304945946, 'timing/model': 1.22809789609164, 'learning_rate/action_model': 1.5500513651198514e-05, 'learning_rate/action_query_proj': 1.5500513651198514e-05, 'learning_rate/base': 3.8751284127996284e-06, 'epoch': 0.67}) +[2026-06-30 07:13:51,654][starVLA.training.train_starvla][INFO] - Step 1551, Loss: {'train/loss': 6.9737097874167375e-06, 'train/grad_norm_pre_clip': 2.2508459091186523, 'timing/data': 0.00020228209905326366, 'timing/model': 1.2303710400592536, 'learning_rate/action_model': 1.5451591471542078e-05, 'learning_rate/action_query_proj': 1.5451591471542078e-05, 'learning_rate/base': 3.8628978678855195e-06, 'epoch': 0.67}) +[2026-06-30 07:14:24,294][starVLA.training.train_starvla][INFO] - Step 1552, Loss: {'train/loss': 0.00012148587120464072, 'train/grad_norm_pre_clip': 2.226294994354248, 'timing/data': 0.00019626086577773094, 'timing/model': 1.2304555401206017, 'learning_rate/action_model': 1.5402759471450494e-05, 'learning_rate/action_query_proj': 1.5402759471450494e-05, 'learning_rate/base': 3.8506898678626235e-06, 'epoch': 0.67}) +[2026-06-30 07:14:56,962][starVLA.training.train_starvla][INFO] - Step 1553, Loss: {'train/loss': 0.0006101648323237896, 'train/grad_norm_pre_clip': 4.001655578613281, 'timing/data': 0.00019880500622093678, 'timing/model': 1.2296954579651356, 'learning_rate/action_model': 1.5354017771411925e-05, 'learning_rate/action_query_proj': 1.5354017771411925e-05, 'learning_rate/base': 3.838504442852981e-06, 'epoch': 0.67}) +[2026-06-30 07:15:29,679][starVLA.training.train_starvla][INFO] - Step 1554, Loss: {'train/loss': 1.668892946327105e-05, 'train/grad_norm_pre_clip': 4.848865509033203, 'timing/data': 0.0001987649593502283, 'timing/model': 1.2287073249462992, 'learning_rate/action_model': 1.5305366491691605e-05, 'learning_rate/action_query_proj': 1.5305366491691605e-05, 'learning_rate/base': 3.826341622922901e-06, 'epoch': 0.67}) +[2026-06-30 07:16:02,367][starVLA.training.train_starvla][INFO] - Step 1555, Loss: {'train/loss': 4.860424451180734e-05, 'train/grad_norm_pre_clip': 5.328364849090576, 'timing/data': 0.0002760209608823061, 'timing/model': 1.2299022730439901, 'learning_rate/action_model': 1.5256805752331777e-05, 'learning_rate/action_query_proj': 1.5256805752331777e-05, 'learning_rate/base': 3.814201438082944e-06, 'epoch': 0.67}) +[2026-06-30 07:16:35,190][starVLA.training.train_starvla][INFO] - Step 1556, Loss: {'train/loss': 0.03751063346862793, 'train/grad_norm_pre_clip': 1.2812405824661255, 'timing/data': 0.00034756679087877274, 'timing/model': 1.229063019854948, 'learning_rate/action_model': 1.5208335673151214e-05, 'learning_rate/action_query_proj': 1.5208335673151214e-05, 'learning_rate/base': 3.8020839182878035e-06, 'epoch': 0.67}) +[2026-06-30 07:17:07,832][starVLA.training.train_starvla][INFO] - Step 1557, Loss: {'train/loss': 0.0005696548032574356, 'train/grad_norm_pre_clip': 3.074786901473999, 'timing/data': 0.0002927829045802355, 'timing/model': 1.23425637697801, 'learning_rate/action_model': 1.5159956373745033e-05, 'learning_rate/action_query_proj': 1.5159956373745033e-05, 'learning_rate/base': 3.7899890934362582e-06, 'epoch': 0.67}) +[2026-06-30 07:17:40,420][starVLA.training.train_starvla][INFO] - Step 1558, Loss: {'train/loss': 0.23728981614112854, 'train/grad_norm_pre_clip': 2.6656572818756104, 'timing/data': 0.00024342001415789127, 'timing/model': 1.227886838838458, 'learning_rate/action_model': 1.5111667973484322e-05, 'learning_rate/action_query_proj': 1.5111667973484322e-05, 'learning_rate/base': 3.7779169933710805e-06, 'epoch': 0.67}) +[2026-06-30 07:18:13,106][starVLA.training.train_starvla][INFO] - Step 1559, Loss: {'train/loss': 0.011233322322368622, 'train/grad_norm_pre_clip': 2.5979127883911133, 'timing/data': 0.0002023419365286827, 'timing/model': 1.229800967965275, 'learning_rate/action_model': 1.5063470591515943e-05, 'learning_rate/action_query_proj': 1.5063470591515943e-05, 'learning_rate/base': 3.765867647878986e-06, 'epoch': 0.67}) +[2026-06-30 07:18:45,774][starVLA.training.train_starvla][INFO] - Step 1560, Loss: {'train/loss': 1.2308265468163881e-05, 'train/grad_norm_pre_clip': 1.9734055995941162, 'timing/data': 0.00026444881223142147, 'timing/model': 1.2331770649179816, 'learning_rate/action_model': 1.501536434676212e-05, 'learning_rate/action_query_proj': 1.501536434676212e-05, 'learning_rate/base': 3.75384108669053e-06, 'epoch': 0.67}) +[2026-06-30 07:19:18,432][starVLA.training.train_starvla][INFO] - Step 1561, Loss: {'train/loss': 0.0005314112058840692, 'train/grad_norm_pre_clip': 2.3392868041992188, 'timing/data': 0.000251515069976449, 'timing/model': 1.2314639200922102, 'learning_rate/action_model': 1.4967349357920245e-05, 'learning_rate/action_query_proj': 1.4967349357920245e-05, 'learning_rate/base': 3.7418373394800614e-06, 'epoch': 0.67}) +[2026-06-30 07:19:50,969][starVLA.training.train_starvla][INFO] - Step 1562, Loss: {'train/loss': 0.07951104640960693, 'train/grad_norm_pre_clip': 2.4745194911956787, 'timing/data': 0.00021038693375885487, 'timing/model': 1.220326503040269, 'learning_rate/action_model': 1.4919425743462507e-05, 'learning_rate/action_query_proj': 1.4919425743462507e-05, 'learning_rate/base': 3.7298564358656267e-06, 'epoch': 0.67}) +[2026-06-30 07:20:23,588][starVLA.training.train_starvla][INFO] - Step 1563, Loss: {'train/loss': 8.434006304014474e-06, 'train/grad_norm_pre_clip': 1.983335256576538, 'timing/data': 0.00019428692758083344, 'timing/model': 1.2282466748729348, 'learning_rate/action_model': 1.4871593621635687e-05, 'learning_rate/action_query_proj': 1.4871593621635687e-05, 'learning_rate/base': 3.7178984054089217e-06, 'epoch': 0.67}) +[2026-06-30 07:20:56,256][starVLA.training.train_starvla][INFO] - Step 1564, Loss: {'train/loss': 0.0018694044556468725, 'train/grad_norm_pre_clip': 3.2871909141540527, 'timing/data': 0.00026019103825092316, 'timing/model': 1.2288533980026841, 'learning_rate/action_model': 1.4823853110460781e-05, 'learning_rate/action_query_proj': 1.4823853110460781e-05, 'learning_rate/base': 3.7059632776151952e-06, 'epoch': 0.67}) +[2026-06-30 07:21:28,955][starVLA.training.train_starvla][INFO] - Step 1565, Loss: {'train/loss': 0.009828120470046997, 'train/grad_norm_pre_clip': 2.247183084487915, 'timing/data': 0.0001905399840325117, 'timing/model': 1.2348086300771683, 'learning_rate/action_model': 1.4776204327732751e-05, 'learning_rate/action_query_proj': 1.4776204327732751e-05, 'learning_rate/base': 3.694051081933188e-06, 'epoch': 0.67}) +[2026-06-30 07:22:01,618][starVLA.training.train_starvla][INFO] - Step 1566, Loss: {'train/loss': 0.07423648238182068, 'train/grad_norm_pre_clip': 2.434293270111084, 'timing/data': 0.0001984238624572754, 'timing/model': 1.2362212231382728, 'learning_rate/action_model': 1.4728647391020222e-05, 'learning_rate/action_query_proj': 1.4728647391020222e-05, 'learning_rate/base': 3.6821618477550556e-06, 'epoch': 0.67}) +[2026-06-30 07:22:34,250][starVLA.training.train_starvla][INFO] - Step 1567, Loss: {'train/loss': 0.015439827926456928, 'train/grad_norm_pre_clip': 3.468777656555176, 'timing/data': 0.000200208043679595, 'timing/model': 1.2286958508193493, 'learning_rate/action_model': 1.4681182417665238e-05, 'learning_rate/action_query_proj': 1.4681182417665238e-05, 'learning_rate/base': 3.6702956044163095e-06, 'epoch': 0.67}) +[2026-06-30 07:23:06,880][starVLA.training.train_starvla][INFO] - Step 1568, Loss: {'train/loss': 0.00038989295717328787, 'train/grad_norm_pre_clip': 4.416195392608643, 'timing/data': 0.00044111208990216255, 'timing/model': 1.2298046899959445, 'learning_rate/action_model': 1.463380952478288e-05, 'learning_rate/action_query_proj': 1.463380952478288e-05, 'learning_rate/base': 3.65845238119572e-06, 'epoch': 0.67}) +[2026-06-30 07:23:39,572][starVLA.training.train_starvla][INFO] - Step 1569, Loss: {'train/loss': 0.9907034039497375, 'train/grad_norm_pre_clip': 2.797128438949585, 'timing/data': 0.0001957891508936882, 'timing/model': 1.2282062920276076, 'learning_rate/action_model': 1.4586528829261054e-05, 'learning_rate/action_query_proj': 1.4586528829261054e-05, 'learning_rate/base': 3.6466322073152635e-06, 'epoch': 0.68}) +[2026-06-30 07:24:12,248][starVLA.training.train_starvla][INFO] - Step 1570, Loss: {'train/loss': 0.02745581790804863, 'train/grad_norm_pre_clip': 2.539963483810425, 'timing/data': 0.0002706910017877817, 'timing/model': 1.227475357009098, 'learning_rate/action_model': 1.4539340447760169e-05, 'learning_rate/action_query_proj': 1.4539340447760169e-05, 'learning_rate/base': 3.634835111940042e-06, 'epoch': 0.68}) +[2026-06-30 07:24:44,907][starVLA.training.train_starvla][INFO] - Step 1571, Loss: {'train/loss': 0.0021744235418736935, 'train/grad_norm_pre_clip': 1.3932089805603027, 'timing/data': 0.0001955090556293726, 'timing/model': 1.2332982930820435, 'learning_rate/action_model': 1.4492244496712897e-05, 'learning_rate/action_query_proj': 1.4492244496712897e-05, 'learning_rate/base': 3.6230611241782243e-06, 'epoch': 0.68}) +[2026-06-30 07:25:17,557][starVLA.training.train_starvla][INFO] - Step 1572, Loss: {'train/loss': 0.5067124366760254, 'train/grad_norm_pre_clip': 2.3395133018493652, 'timing/data': 0.00019890489056706429, 'timing/model': 1.2302951260935515, 'learning_rate/action_model': 1.444524109232377e-05, 'learning_rate/action_query_proj': 1.444524109232377e-05, 'learning_rate/base': 3.6113102730809424e-06, 'epoch': 0.68}) +[2026-06-30 07:25:50,217][starVLA.training.train_starvla][INFO] - Step 1573, Loss: {'train/loss': 0.011548415757715702, 'train/grad_norm_pre_clip': 3.146756887435913, 'timing/data': 0.0002517548855394125, 'timing/model': 1.2296696279663593, 'learning_rate/action_model': 1.439833035056905e-05, 'learning_rate/action_query_proj': 1.439833035056905e-05, 'learning_rate/base': 3.5995825876422626e-06, 'epoch': 0.68}) +[2026-06-30 07:26:22,894][starVLA.training.train_starvla][INFO] - Step 1574, Loss: {'train/loss': 0.015278800390660763, 'train/grad_norm_pre_clip': 2.037095785140991, 'timing/data': 0.0002020911779254675, 'timing/model': 1.229551682015881, 'learning_rate/action_model': 1.4351512387196308e-05, 'learning_rate/action_query_proj': 1.4351512387196308e-05, 'learning_rate/base': 3.587878096799077e-06, 'epoch': 0.68}) +[2026-06-30 07:26:55,544][starVLA.training.train_starvla][INFO] - Step 1575, Loss: {'train/loss': 0.04696502536535263, 'train/grad_norm_pre_clip': 2.0339736938476562, 'timing/data': 0.00020405510440468788, 'timing/model': 1.2358563779853284, 'learning_rate/action_model': 1.4304787317724252e-05, 'learning_rate/action_query_proj': 1.4304787317724252e-05, 'learning_rate/base': 3.576196829431063e-06, 'epoch': 0.68}) +[2026-06-30 07:27:28,212][starVLA.training.train_starvla][INFO] - Step 1576, Loss: {'train/loss': 1.0281733921146952e-05, 'train/grad_norm_pre_clip': 2.9498167037963867, 'timing/data': 0.00019558914937078953, 'timing/model': 1.2311789940577, 'learning_rate/action_model': 1.425815525744229e-05, 'learning_rate/action_query_proj': 1.425815525744229e-05, 'learning_rate/base': 3.5645388143605725e-06, 'epoch': 0.68}) +[2026-06-30 07:28:00,892][starVLA.training.train_starvla][INFO] - Step 1577, Loss: {'train/loss': 0.08867358416318893, 'train/grad_norm_pre_clip': 2.15030837059021, 'timing/data': 0.00020578899420797825, 'timing/model': 1.2337382659316063, 'learning_rate/action_model': 1.4211616321410436e-05, 'learning_rate/action_query_proj': 1.4211616321410436e-05, 'learning_rate/base': 3.552904080352609e-06, 'epoch': 0.68}) +[2026-06-30 07:28:33,572][starVLA.training.train_starvla][INFO] - Step 1578, Loss: {'train/loss': 9.387661521031987e-06, 'train/grad_norm_pre_clip': 2.346357822418213, 'timing/data': 0.00044317683205008507, 'timing/model': 1.2296877990011126, 'learning_rate/action_model': 1.4165170624458881e-05, 'learning_rate/action_query_proj': 1.4165170624458881e-05, 'learning_rate/base': 3.5412926561147204e-06, 'epoch': 0.68}) +[2026-06-30 07:29:06,256][starVLA.training.train_starvla][INFO] - Step 1579, Loss: {'train/loss': 0.019098972901701927, 'train/grad_norm_pre_clip': 1.905137300491333, 'timing/data': 0.00020141992717981339, 'timing/model': 1.2318549761548638, 'learning_rate/action_model': 1.411881828118776e-05, 'learning_rate/action_query_proj': 1.411881828118776e-05, 'learning_rate/base': 3.52970457029694e-06, 'epoch': 0.68}) +[2026-06-30 07:29:38,904][starVLA.training.train_starvla][INFO] - Step 1580, Loss: {'train/loss': 0.1946604698896408, 'train/grad_norm_pre_clip': 3.6470859050750732, 'timing/data': 0.00029717106372117996, 'timing/model': 1.228481702040881, 'learning_rate/action_model': 1.407255940596686e-05, 'learning_rate/action_query_proj': 1.407255940596686e-05, 'learning_rate/base': 3.518139851491715e-06, 'epoch': 0.68}) +[2026-06-30 07:30:11,580][starVLA.training.train_starvla][INFO] - Step 1581, Loss: {'train/loss': 0.0002857146319001913, 'train/grad_norm_pre_clip': 1.9169937372207642, 'timing/data': 0.0002169900108128786, 'timing/model': 1.226268167840317, 'learning_rate/action_model': 1.4026394112935381e-05, 'learning_rate/action_query_proj': 1.4026394112935381e-05, 'learning_rate/base': 3.5065985282338452e-06, 'epoch': 0.68}) +[2026-06-30 07:30:44,168][starVLA.training.train_starvla][INFO] - Step 1582, Loss: {'train/loss': 0.003369538579136133, 'train/grad_norm_pre_clip': 4.439128875732422, 'timing/data': 0.00026804604567587376, 'timing/model': 1.2260373851750046, 'learning_rate/action_model': 1.3980322516001581e-05, 'learning_rate/action_query_proj': 1.3980322516001581e-05, 'learning_rate/base': 3.4950806290003953e-06, 'epoch': 0.68}) +[2026-06-30 07:31:16,851][starVLA.training.train_starvla][INFO] - Step 1583, Loss: {'train/loss': 0.8968007564544678, 'train/grad_norm_pre_clip': 2.5396337509155273, 'timing/data': 0.00021942402236163616, 'timing/model': 1.2330488301813602, 'learning_rate/action_model': 1.3934344728842556e-05, 'learning_rate/action_query_proj': 1.3934344728842556e-05, 'learning_rate/base': 3.483586182210639e-06, 'epoch': 0.68}) +[2026-06-30 07:31:49,511][starVLA.training.train_starvla][INFO] - Step 1584, Loss: {'train/loss': 0.22785021364688873, 'train/grad_norm_pre_clip': 1.934855580329895, 'timing/data': 0.00025072298012673855, 'timing/model': 1.2296222520526499, 'learning_rate/action_model': 1.3888460864903902e-05, 'learning_rate/action_query_proj': 1.3888460864903902e-05, 'learning_rate/base': 3.4721152162259756e-06, 'epoch': 0.68}) +[2026-06-30 07:32:22,188][starVLA.training.train_starvla][INFO] - Step 1585, Loss: {'train/loss': 0.00020085378491785377, 'train/grad_norm_pre_clip': 1.6912935972213745, 'timing/data': 0.00029758084565401077, 'timing/model': 1.2290642280131578, 'learning_rate/action_model': 1.3842671037399518e-05, 'learning_rate/action_query_proj': 1.3842671037399518e-05, 'learning_rate/base': 3.4606677593498795e-06, 'epoch': 0.68}) +[2026-06-30 07:32:54,822][starVLA.training.train_starvla][INFO] - Step 1586, Loss: {'train/loss': 0.08690037578344345, 'train/grad_norm_pre_clip': 2.7771992683410645, 'timing/data': 0.00020301295444369316, 'timing/model': 1.2325780971441418, 'learning_rate/action_model': 1.379697535931124e-05, 'learning_rate/action_query_proj': 1.379697535931124e-05, 'learning_rate/base': 3.44924383982781e-06, 'epoch': 0.68}) +[2026-06-30 07:33:27,476][starVLA.training.train_starvla][INFO] - Step 1587, Loss: {'train/loss': 0.020290138199925423, 'train/grad_norm_pre_clip': 1.584197998046875, 'timing/data': 0.00020617898553609848, 'timing/model': 1.2350339209660888, 'learning_rate/action_model': 1.3751373943388613e-05, 'learning_rate/action_query_proj': 1.3751373943388613e-05, 'learning_rate/base': 3.4378434858471532e-06, 'epoch': 0.68}) +[2026-06-30 07:34:00,117][starVLA.training.train_starvla][INFO] - Step 1588, Loss: {'train/loss': 1.2904294635518454e-05, 'train/grad_norm_pre_clip': 2.567232847213745, 'timing/data': 0.00020523695275187492, 'timing/model': 1.2190018759574741, 'learning_rate/action_model': 1.3705866902148587e-05, 'learning_rate/action_query_proj': 1.3705866902148587e-05, 'learning_rate/base': 3.426466725537147e-06, 'epoch': 0.68}) +[2026-06-30 07:34:32,830][starVLA.training.train_starvla][INFO] - Step 1589, Loss: {'train/loss': 6.511441461043432e-05, 'train/grad_norm_pre_clip': 2.5439982414245605, 'timing/data': 0.00023098615929484367, 'timing/model': 1.2285300251096487, 'learning_rate/action_model': 1.3660454347875298e-05, 'learning_rate/action_query_proj': 1.3660454347875298e-05, 'learning_rate/base': 3.4151135869688245e-06, 'epoch': 0.68}) +[2026-06-30 07:35:05,483][starVLA.training.train_starvla][INFO] - Step 1590, Loss: {'train/loss': 0.00014542842109221965, 'train/grad_norm_pre_clip': 2.098109245300293, 'timing/data': 0.00019284384325146675, 'timing/model': 1.2307996519375592, 'learning_rate/action_model': 1.361513639261965e-05, 'learning_rate/action_query_proj': 1.361513639261965e-05, 'learning_rate/base': 3.4037840981549126e-06, 'epoch': 0.68}) +[2026-06-30 07:35:38,123][starVLA.training.train_starvla][INFO] - Step 1591, Loss: {'train/loss': 0.04976874217391014, 'train/grad_norm_pre_clip': 1.455278992652893, 'timing/data': 0.0002049671020358801, 'timing/model': 1.2304545158986002, 'learning_rate/action_model': 1.3569913148199249e-05, 'learning_rate/action_query_proj': 1.3569913148199249e-05, 'learning_rate/base': 3.392478287049812e-06, 'epoch': 0.68}) +[2026-06-30 07:36:10,777][starVLA.training.train_starvla][INFO] - Step 1592, Loss: {'train/loss': 0.0028948618564754725, 'train/grad_norm_pre_clip': 1.443039059638977, 'timing/data': 0.0005216849967837334, 'timing/model': 1.2308107349090278, 'learning_rate/action_model': 1.3524784726197915e-05, 'learning_rate/action_query_proj': 1.3524784726197915e-05, 'learning_rate/base': 3.3811961815494787e-06, 'epoch': 0.69}) +[2026-06-30 07:36:43,437][starVLA.training.train_starvla][INFO] - Step 1593, Loss: {'train/loss': 0.019643612205982208, 'train/grad_norm_pre_clip': 1.9814502000808716, 'timing/data': 0.00020196102559566498, 'timing/model': 1.2412612771149725, 'learning_rate/action_model': 1.3479751237965587e-05, 'learning_rate/action_query_proj': 1.3479751237965587e-05, 'learning_rate/base': 3.369937809491397e-06, 'epoch': 0.69}) +[2026-06-30 07:37:16,076][starVLA.training.train_starvla][INFO] - Step 1594, Loss: {'train/loss': 1.5437401089002378e-05, 'train/grad_norm_pre_clip': 2.064279079437256, 'timing/data': 0.00020265299826860428, 'timing/model': 1.2196884951554239, 'learning_rate/action_model': 1.3434812794617874e-05, 'learning_rate/action_query_proj': 1.3434812794617874e-05, 'learning_rate/base': 3.3587031986544686e-06, 'epoch': 0.69}) +[2026-06-30 07:37:48,703][starVLA.training.train_starvla][INFO] - Step 1595, Loss: {'train/loss': 0.08170851320028305, 'train/grad_norm_pre_clip': 1.3893890380859375, 'timing/data': 0.00039120810106396675, 'timing/model': 1.2327756138984114, 'learning_rate/action_model': 1.3389969507035944e-05, 'learning_rate/action_query_proj': 1.3389969507035944e-05, 'learning_rate/base': 3.347492376758986e-06, 'epoch': 0.69}) +[2026-06-30 07:38:21,366][starVLA.training.train_starvla][INFO] - Step 1596, Loss: {'train/loss': 0.05025826394557953, 'train/grad_norm_pre_clip': 4.2210259437561035, 'timing/data': 0.00020771194249391556, 'timing/model': 1.2281068959273398, 'learning_rate/action_model': 1.334522148586615e-05, 'learning_rate/action_query_proj': 1.334522148586615e-05, 'learning_rate/base': 3.3363053714665375e-06, 'epoch': 0.69}) +[2026-06-30 07:38:54,003][starVLA.training.train_starvla][INFO] - Step 1597, Loss: {'train/loss': 0.0001575324422447011, 'train/grad_norm_pre_clip': 2.3233938217163086, 'timing/data': 0.00025864806957542896, 'timing/model': 1.231125770136714, 'learning_rate/action_model': 1.3300568841519778e-05, 'learning_rate/action_query_proj': 1.3300568841519778e-05, 'learning_rate/base': 3.3251422103799444e-06, 'epoch': 0.69}) +[2026-06-30 07:39:26,660][starVLA.training.train_starvla][INFO] - Step 1598, Loss: {'train/loss': 0.0010288219200447202, 'train/grad_norm_pre_clip': 1.605672001838684, 'timing/data': 0.00019958592019975185, 'timing/model': 1.2272394190076739, 'learning_rate/action_model': 1.3256011684172787e-05, 'learning_rate/action_query_proj': 1.3256011684172787e-05, 'learning_rate/base': 3.314002921043197e-06, 'epoch': 0.69}) +[2026-06-30 07:39:59,309][starVLA.training.train_starvla][INFO] - Step 1599, Loss: {'train/loss': 0.14301583170890808, 'train/grad_norm_pre_clip': 2.722356081008911, 'timing/data': 0.0001917220652103424, 'timing/model': 1.2298427920322865, 'learning_rate/action_model': 1.3211550123765543e-05, 'learning_rate/action_query_proj': 1.3211550123765543e-05, 'learning_rate/base': 3.3028875309413857e-06, 'epoch': 0.69}) +[2026-06-30 07:40:31,980][starVLA.training.train_starvla][INFO] - Step 1600, Loss: {'train/loss': 0.02242494560778141, 'train/grad_norm_pre_clip': 3.2169389724731445, 'timing/data': 0.00020213215611875057, 'timing/model': 1.2329491740092635, 'learning_rate/action_model': 1.3167184270002527e-05, 'learning_rate/action_query_proj': 1.3167184270002527e-05, 'learning_rate/base': 3.291796067500632e-06, 'epoch': 0.69}) +[2026-06-30 07:41:04,644][starVLA.training.train_starvla][INFO] - Step 1601, Loss: {'train/loss': 8.434016308456194e-06, 'train/grad_norm_pre_clip': 1.8635032176971436, 'timing/data': 0.00019391602836549282, 'timing/model': 1.2327406730037183, 'learning_rate/action_model': 1.3122914232352069e-05, 'learning_rate/action_query_proj': 1.3122914232352069e-05, 'learning_rate/base': 3.280728558088017e-06, 'epoch': 0.69}) +[2026-06-30 07:41:37,338][starVLA.training.train_starvla][INFO] - Step 1602, Loss: {'train/loss': 0.0016300927381962538, 'train/grad_norm_pre_clip': 2.745234727859497, 'timing/data': 0.0002670029643923044, 'timing/model': 1.2286177568603307, 'learning_rate/action_model': 1.3078740120046071e-05, 'learning_rate/action_query_proj': 1.3078740120046071e-05, 'learning_rate/base': 3.269685030011518e-06, 'epoch': 0.69}) +[2026-06-30 07:42:09,983][starVLA.training.train_starvla][INFO] - Step 1603, Loss: {'train/loss': 0.26618778705596924, 'train/grad_norm_pre_clip': 3.463574171066284, 'timing/data': 0.00028792303055524826, 'timing/model': 1.2344891580287367, 'learning_rate/action_model': 1.3034662042079792e-05, 'learning_rate/action_query_proj': 1.3034662042079792e-05, 'learning_rate/base': 3.258665510519948e-06, 'epoch': 0.69}) +[2026-06-30 07:42:42,690][starVLA.training.train_starvla][INFO] - Step 1604, Loss: {'train/loss': 9.35786829359131e-06, 'train/grad_norm_pre_clip': 4.877245903015137, 'timing/data': 0.00019910605624318123, 'timing/model': 1.2338397959247231, 'learning_rate/action_model': 1.2990680107211489e-05, 'learning_rate/action_query_proj': 1.2990680107211489e-05, 'learning_rate/base': 3.247670026802872e-06, 'epoch': 0.69}) +[2026-06-30 07:43:15,382][starVLA.training.train_starvla][INFO] - Step 1605, Loss: {'train/loss': 0.03911779820919037, 'train/grad_norm_pre_clip': 2.726090431213379, 'timing/data': 0.00025797588750720024, 'timing/model': 1.23345676693134, 'learning_rate/action_model': 1.2946794423962228e-05, 'learning_rate/action_query_proj': 1.2946794423962228e-05, 'learning_rate/base': 3.236698605990557e-06, 'epoch': 0.69}) +[2026-06-30 07:43:48,055][starVLA.training.train_starvla][INFO] - Step 1606, Loss: {'train/loss': 0.00040580006316304207, 'train/grad_norm_pre_clip': 2.1978256702423096, 'timing/data': 0.0002835050690919161, 'timing/model': 1.2328599751926959, 'learning_rate/action_model': 1.2903005100615544e-05, 'learning_rate/action_query_proj': 1.2903005100615544e-05, 'learning_rate/base': 3.225751275153886e-06, 'epoch': 0.69}) +[2026-06-30 07:44:20,745][starVLA.training.train_starvla][INFO] - Step 1607, Loss: {'train/loss': 0.0012488848296925426, 'train/grad_norm_pre_clip': 2.236194133758545, 'timing/data': 0.00028540915809571743, 'timing/model': 1.2338486749213189, 'learning_rate/action_model': 1.285931224521728e-05, 'learning_rate/action_query_proj': 1.285931224521728e-05, 'learning_rate/base': 3.21482806130432e-06, 'epoch': 0.69}) +[2026-06-30 07:44:53,625][starVLA.training.train_starvla][INFO] - Step 1608, Loss: {'train/loss': 0.001189632574096322, 'train/grad_norm_pre_clip': 2.363774538040161, 'timing/data': 0.0002902778796851635, 'timing/model': 1.2353973318822682, 'learning_rate/action_model': 1.281571596557517e-05, 'learning_rate/action_query_proj': 1.281571596557517e-05, 'learning_rate/base': 3.2039289913937926e-06, 'epoch': 0.69}) +[2026-06-30 07:45:26,297][starVLA.training.train_starvla][INFO] - Step 1609, Loss: {'train/loss': 0.053637195378541946, 'train/grad_norm_pre_clip': 2.7316372394561768, 'timing/data': 0.00037950603291392326, 'timing/model': 1.2312003599945456, 'learning_rate/action_model': 1.2772216369258733e-05, 'learning_rate/action_query_proj': 1.2772216369258733e-05, 'learning_rate/base': 3.193054092314683e-06, 'epoch': 0.69}) +[2026-06-30 07:45:58,952][starVLA.training.train_starvla][INFO] - Step 1610, Loss: {'train/loss': 1.0043311704066582e-05, 'train/grad_norm_pre_clip': 3.0228774547576904, 'timing/data': 0.0003685259725898504, 'timing/model': 1.232310689985752, 'learning_rate/action_model': 1.2728813563598865e-05, 'learning_rate/action_query_proj': 1.2728813563598865e-05, 'learning_rate/base': 3.1822033908997163e-06, 'epoch': 0.69}) +[2026-06-30 07:46:31,602][starVLA.training.train_starvla][INFO] - Step 1611, Loss: {'train/loss': 4.473206354305148e-05, 'train/grad_norm_pre_clip': 2.478130340576172, 'timing/data': 0.00035892706364393234, 'timing/model': 1.2312104508746415, 'learning_rate/action_model': 1.2685507655687732e-05, 'learning_rate/action_query_proj': 1.2685507655687732e-05, 'learning_rate/base': 3.171376913921933e-06, 'epoch': 0.69}) +[2026-06-30 07:47:04,259][starVLA.training.train_starvla][INFO] - Step 1612, Loss: {'train/loss': 6.928775837877765e-05, 'train/grad_norm_pre_clip': 2.72894549369812, 'timing/data': 0.0002569539938122034, 'timing/model': 1.2326255049556494, 'learning_rate/action_model': 1.2642298752378285e-05, 'learning_rate/action_query_proj': 1.2642298752378285e-05, 'learning_rate/base': 3.160574688094571e-06, 'epoch': 0.69}) +[2026-06-30 07:47:36,922][starVLA.training.train_starvla][INFO] - Step 1613, Loss: {'train/loss': 0.0008614988764747977, 'train/grad_norm_pre_clip': 2.341914176940918, 'timing/data': 0.0003976188600063324, 'timing/model': 1.2285814969800413, 'learning_rate/action_model': 1.259918696028425e-05, 'learning_rate/action_query_proj': 1.259918696028425e-05, 'learning_rate/base': 3.1497967400710626e-06, 'epoch': 0.69}) +[2026-06-30 07:48:09,612][starVLA.training.train_starvla][INFO] - Step 1614, Loss: {'train/loss': 4.2048613977385685e-05, 'train/grad_norm_pre_clip': 2.029541015625, 'timing/data': 0.00020041782408952713, 'timing/model': 1.2299662278965116, 'learning_rate/action_model': 1.2556172385779674e-05, 'learning_rate/action_query_proj': 1.2556172385779674e-05, 'learning_rate/base': 3.1390430964449184e-06, 'epoch': 0.69}) +[2026-06-30 07:48:42,291][starVLA.training.train_starvla][INFO] - Step 1615, Loss: {'train/loss': 0.3074752688407898, 'train/grad_norm_pre_clip': 3.383467674255371, 'timing/data': 0.0002014990895986557, 'timing/model': 1.2331454150844365, 'learning_rate/action_model': 1.2513255134998737e-05, 'learning_rate/action_query_proj': 1.2513255134998737e-05, 'learning_rate/base': 3.128313783749684e-06, 'epoch': 0.69}) +[2026-06-30 07:49:14,937][starVLA.training.train_starvla][INFO] - Step 1616, Loss: {'train/loss': 0.0005967022152617574, 'train/grad_norm_pre_clip': 1.970967173576355, 'timing/data': 0.0002467741724103689, 'timing/model': 1.2293404429219663, 'learning_rate/action_model': 1.2470435313835483e-05, 'learning_rate/action_query_proj': 1.2470435313835483e-05, 'learning_rate/base': 3.1176088284588706e-06, 'epoch': 0.7}) +[2026-06-30 07:49:47,614][starVLA.training.train_starvla][INFO] - Step 1617, Loss: {'train/loss': 0.0005954018561169505, 'train/grad_norm_pre_clip': 3.326399564743042, 'timing/data': 0.0002465648576617241, 'timing/model': 1.2205305418465286, 'learning_rate/action_model': 1.2427713027943582e-05, 'learning_rate/action_query_proj': 1.2427713027943582e-05, 'learning_rate/base': 3.1069282569858954e-06, 'epoch': 0.7}) +[2026-06-30 07:50:20,279][starVLA.training.train_starvla][INFO] - Step 1618, Loss: {'train/loss': 0.22777312994003296, 'train/grad_norm_pre_clip': 1.7497375011444092, 'timing/data': 0.00024518207646906376, 'timing/model': 1.2309131510555744, 'learning_rate/action_model': 1.238508838273602e-05, 'learning_rate/action_query_proj': 1.238508838273602e-05, 'learning_rate/base': 3.096272095684005e-06, 'epoch': 0.7}) +[2026-06-30 07:50:52,900][starVLA.training.train_starvla][INFO] - Step 1619, Loss: {'train/loss': 0.05977890267968178, 'train/grad_norm_pre_clip': 1.7953870296478271, 'timing/data': 0.00019548810087144375, 'timing/model': 1.233522695954889, 'learning_rate/action_model': 1.2342561483384875e-05, 'learning_rate/action_query_proj': 1.2342561483384875e-05, 'learning_rate/base': 3.085640370846219e-06, 'epoch': 0.7}) +[2026-06-30 07:51:25,519][starVLA.training.train_starvla][INFO] - Step 1620, Loss: {'train/loss': 0.06804933398962021, 'train/grad_norm_pre_clip': 2.549638032913208, 'timing/data': 0.0003140820190310478, 'timing/model': 1.232298678951338, 'learning_rate/action_model': 1.2300132434821026e-05, 'learning_rate/action_query_proj': 1.2300132434821026e-05, 'learning_rate/base': 3.0750331087052566e-06, 'epoch': 0.7}) +[2026-06-30 07:51:58,151][starVLA.training.train_starvla][INFO] - Step 1621, Loss: {'train/loss': 0.004883202724158764, 'train/grad_norm_pre_clip': 1.4784464836120605, 'timing/data': 0.00019646994769573212, 'timing/model': 1.229357826989144, 'learning_rate/action_model': 1.2257801341733958e-05, 'learning_rate/action_query_proj': 1.2257801341733958e-05, 'learning_rate/base': 3.0644503354334894e-06, 'epoch': 0.7}) +[2026-06-30 07:52:30,808][starVLA.training.train_starvla][INFO] - Step 1622, Loss: {'train/loss': 0.39336538314819336, 'train/grad_norm_pre_clip': 1.7870054244995117, 'timing/data': 0.0005044310819357634, 'timing/model': 1.2272770758718252, 'learning_rate/action_model': 1.2215568308571429e-05, 'learning_rate/action_query_proj': 1.2215568308571429e-05, 'learning_rate/base': 3.0538920771428572e-06, 'epoch': 0.7}) +[2026-06-30 07:53:03,453][starVLA.training.train_starvla][INFO] - Step 1623, Loss: {'train/loss': 1.347050056210719e-05, 'train/grad_norm_pre_clip': 1.7626800537109375, 'timing/data': 0.0002877830993384123, 'timing/model': 1.2294799690134823, 'learning_rate/action_model': 1.2173433439539243e-05, 'learning_rate/action_query_proj': 1.2173433439539243e-05, 'learning_rate/base': 3.043358359884811e-06, 'epoch': 0.7}) +[2026-06-30 07:53:36,048][starVLA.training.train_starvla][INFO] - Step 1624, Loss: {'train/loss': 0.026985125616192818, 'train/grad_norm_pre_clip': 2.041757583618164, 'timing/data': 0.000203313073143363, 'timing/model': 1.2256861210335046, 'learning_rate/action_model': 1.2131396838600986e-05, 'learning_rate/action_query_proj': 1.2131396838600986e-05, 'learning_rate/base': 3.0328492096502465e-06, 'epoch': 0.7}) +[2026-06-30 07:54:08,727][starVLA.training.train_starvla][INFO] - Step 1625, Loss: {'train/loss': 0.09072209149599075, 'train/grad_norm_pre_clip': 1.649798035621643, 'timing/data': 0.0002903868444263935, 'timing/model': 1.2328948720823973, 'learning_rate/action_model': 1.2089458609477822e-05, 'learning_rate/action_query_proj': 1.2089458609477822e-05, 'learning_rate/base': 3.0223646523694556e-06, 'epoch': 0.7}) +[2026-06-30 07:54:41,344][starVLA.training.train_starvla][INFO] - Step 1626, Loss: {'train/loss': 0.00015935339615680277, 'train/grad_norm_pre_clip': 1.6937538385391235, 'timing/data': 0.00019796309061348438, 'timing/model': 1.2327890121378005, 'learning_rate/action_model': 1.2047618855648162e-05, 'learning_rate/action_query_proj': 1.2047618855648162e-05, 'learning_rate/base': 3.0119047139120406e-06, 'epoch': 0.7}) +[2026-06-30 07:55:13,978][starVLA.training.train_starvla][INFO] - Step 1627, Loss: {'train/loss': 0.005886138416826725, 'train/grad_norm_pre_clip': 3.467210292816162, 'timing/data': 0.0004613990895450115, 'timing/model': 1.2321748360991478, 'learning_rate/action_model': 1.2005877680347406e-05, 'learning_rate/action_query_proj': 1.2005877680347406e-05, 'learning_rate/base': 3.0014694200868514e-06, 'epoch': 0.7}) +[2026-06-30 07:55:46,659][starVLA.training.train_starvla][INFO] - Step 1628, Loss: {'train/loss': 0.019703581929206848, 'train/grad_norm_pre_clip': 2.5883090496063232, 'timing/data': 0.0001945560798048973, 'timing/model': 1.2290156970266253, 'learning_rate/action_model': 1.1964235186567784e-05, 'learning_rate/action_query_proj': 1.1964235186567784e-05, 'learning_rate/base': 2.991058796641946e-06, 'epoch': 0.7}) +[2026-06-30 07:56:19,318][starVLA.training.train_starvla][INFO] - Step 1629, Loss: {'train/loss': 0.16877588629722595, 'train/grad_norm_pre_clip': 2.094860315322876, 'timing/data': 0.00029672891832888126, 'timing/model': 1.2312065560836345, 'learning_rate/action_model': 1.1922691477058005e-05, 'learning_rate/action_query_proj': 1.1922691477058005e-05, 'learning_rate/base': 2.980672869264501e-06, 'epoch': 0.7}) +[2026-06-30 07:56:51,961][starVLA.training.train_starvla][INFO] - Step 1630, Loss: {'train/loss': 0.008514861576259136, 'train/grad_norm_pre_clip': 2.3159146308898926, 'timing/data': 0.00020630890503525734, 'timing/model': 1.2313184309750795, 'learning_rate/action_model': 1.1881246654323038e-05, 'learning_rate/action_query_proj': 1.1881246654323038e-05, 'learning_rate/base': 2.9703116635807595e-06, 'epoch': 0.7}) +[2026-06-30 07:57:24,671][starVLA.training.train_starvla][INFO] - Step 1631, Loss: {'train/loss': 0.0792175680398941, 'train/grad_norm_pre_clip': 3.047175884246826, 'timing/data': 0.00021256995387375355, 'timing/model': 1.2331419182009995, 'learning_rate/action_model': 1.1839900820623847e-05, 'learning_rate/action_query_proj': 1.1839900820623847e-05, 'learning_rate/base': 2.959975205155962e-06, 'epoch': 0.7}) +[2026-06-30 07:57:57,360][starVLA.training.train_starvla][INFO] - Step 1632, Loss: {'train/loss': 0.21846893429756165, 'train/grad_norm_pre_clip': 1.9646836519241333, 'timing/data': 0.0002666930668056011, 'timing/model': 1.2298917358275503, 'learning_rate/action_model': 1.1798654077977201e-05, 'learning_rate/action_query_proj': 1.1798654077977201e-05, 'learning_rate/base': 2.9496635194943003e-06, 'epoch': 0.7}) +[2026-06-30 07:58:30,025][starVLA.training.train_starvla][INFO] - Step 1633, Loss: {'train/loss': 0.003671437269076705, 'train/grad_norm_pre_clip': 2.0912373065948486, 'timing/data': 0.00019996706396341324, 'timing/model': 1.2339204389136285, 'learning_rate/action_model': 1.1757506528155318e-05, 'learning_rate/action_query_proj': 1.1757506528155318e-05, 'learning_rate/base': 2.9393766320388294e-06, 'epoch': 0.7}) +[2026-06-30 07:59:02,703][starVLA.training.train_starvla][INFO] - Step 1634, Loss: {'train/loss': 0.00019888667156919837, 'train/grad_norm_pre_clip': 3.9573442935943604, 'timing/data': 0.00024110404774546623, 'timing/model': 1.2353752178605646, 'learning_rate/action_model': 1.1716458272685685e-05, 'learning_rate/action_query_proj': 1.1716458272685685e-05, 'learning_rate/base': 2.929114568171421e-06, 'epoch': 0.7}) +[2026-06-30 07:59:35,394][starVLA.training.train_starvla][INFO] - Step 1635, Loss: {'train/loss': 6.020048203936312e-06, 'train/grad_norm_pre_clip': 1.7706085443496704, 'timing/data': 0.00020126905292272568, 'timing/model': 1.2329202610999346, 'learning_rate/action_model': 1.1675509412850789e-05, 'learning_rate/action_query_proj': 1.1675509412850789e-05, 'learning_rate/base': 2.918877353212697e-06, 'epoch': 0.7}) +[2026-06-30 08:00:08,058][starVLA.training.train_starvla][INFO] - Step 1636, Loss: {'train/loss': 0.019960744306445122, 'train/grad_norm_pre_clip': 2.1180946826934814, 'timing/data': 0.00028976588509976864, 'timing/model': 1.2265973170287907, 'learning_rate/action_model': 1.1634660049687886e-05, 'learning_rate/action_query_proj': 1.1634660049687886e-05, 'learning_rate/base': 2.9086650124219714e-06, 'epoch': 0.7}) +[2026-06-30 08:00:40,726][starVLA.training.train_starvla][INFO] - Step 1637, Loss: {'train/loss': 0.00014146538160275668, 'train/grad_norm_pre_clip': 1.8013938665390015, 'timing/data': 0.00034981989301741123, 'timing/model': 1.2285870318301022, 'learning_rate/action_model': 1.159391028398871e-05, 'learning_rate/action_query_proj': 1.159391028398871e-05, 'learning_rate/base': 2.8984775709971776e-06, 'epoch': 0.7}) +[2026-06-30 08:01:13,412][starVLA.training.train_starvla][INFO] - Step 1638, Loss: {'train/loss': 4.917370461043902e-06, 'train/grad_norm_pre_clip': 2.4057247638702393, 'timing/data': 0.0002055480144917965, 'timing/model': 1.2339222619775683, 'learning_rate/action_model': 1.155326021629926e-05, 'learning_rate/action_query_proj': 1.155326021629926e-05, 'learning_rate/base': 2.888315054074815e-06, 'epoch': 0.7}) +[2026-06-30 08:01:46,092][starVLA.training.train_starvla][INFO] - Step 1639, Loss: {'train/loss': 0.001171996584162116, 'train/grad_norm_pre_clip': 1.4155751466751099, 'timing/data': 0.0002984830643981695, 'timing/model': 1.230170605937019, 'learning_rate/action_model': 1.1512709946919516e-05, 'learning_rate/action_query_proj': 1.1512709946919516e-05, 'learning_rate/base': 2.878177486729879e-06, 'epoch': 0.71}) +[2026-06-30 08:02:18,784][starVLA.training.train_starvla][INFO] - Step 1640, Loss: {'train/loss': 0.015579291619360447, 'train/grad_norm_pre_clip': 2.415402412414551, 'timing/data': 0.00019953702576458454, 'timing/model': 1.2306813430041075, 'learning_rate/action_model': 1.1472259575903286e-05, 'learning_rate/action_query_proj': 1.1472259575903286e-05, 'learning_rate/base': 2.8680648939758215e-06, 'epoch': 0.71}) +[2026-06-30 08:02:51,466][starVLA.training.train_starvla][INFO] - Step 1641, Loss: {'train/loss': 0.0008527635945938528, 'train/grad_norm_pre_clip': 2.6855554580688477, 'timing/data': 0.0002886638976633549, 'timing/model': 1.2399972099810839, 'learning_rate/action_model': 1.1431909203057787e-05, 'learning_rate/action_query_proj': 1.1431909203057787e-05, 'learning_rate/base': 2.8579773007644466e-06, 'epoch': 0.71}) +[2026-06-30 08:03:24,130][starVLA.training.train_starvla][INFO] - Step 1642, Loss: {'train/loss': 6.109455171099398e-06, 'train/grad_norm_pre_clip': 2.4649930000305176, 'timing/data': 0.00027193292044103146, 'timing/model': 1.2320639609824866, 'learning_rate/action_model': 1.1391658927943579e-05, 'learning_rate/action_query_proj': 1.1391658927943579e-05, 'learning_rate/base': 2.8479147319858947e-06, 'epoch': 0.71}) +[2026-06-30 08:03:56,755][starVLA.training.train_starvla][INFO] - Step 1643, Loss: {'train/loss': 3.27519592246972e-05, 'train/grad_norm_pre_clip': 2.3560116291046143, 'timing/data': 0.00020108907483518124, 'timing/model': 1.2177049920428544, 'learning_rate/action_model': 1.1351508849874204e-05, 'learning_rate/action_query_proj': 1.1351508849874204e-05, 'learning_rate/base': 2.837877212468551e-06, 'epoch': 0.71}) +[2026-06-30 08:04:29,443][starVLA.training.train_starvla][INFO] - Step 1644, Loss: {'train/loss': 0.03829789534211159, 'train/grad_norm_pre_clip': 2.1067111492156982, 'timing/data': 0.00031345104798674583, 'timing/model': 1.2299156098160893, 'learning_rate/action_model': 1.1311459067916014e-05, 'learning_rate/action_query_proj': 1.1311459067916014e-05, 'learning_rate/base': 2.8278647669790036e-06, 'epoch': 0.71}) +[2026-06-30 08:05:02,117][starVLA.training.train_starvla][INFO] - Step 1645, Loss: {'train/loss': 0.0004801371833309531, 'train/grad_norm_pre_clip': 1.7915514707565308, 'timing/data': 0.00019859406165778637, 'timing/model': 1.2270862970035523, 'learning_rate/action_model': 1.1271509680887828e-05, 'learning_rate/action_query_proj': 1.1271509680887828e-05, 'learning_rate/base': 2.817877420221957e-06, 'epoch': 0.71}) +[2026-06-30 08:05:34,796][starVLA.training.train_starvla][INFO] - Step 1646, Loss: {'train/loss': 0.0005287144449539483, 'train/grad_norm_pre_clip': 3.8115410804748535, 'timing/data': 0.00044095120392739773, 'timing/model': 1.2297313560266048, 'learning_rate/action_model': 1.123166078736081e-05, 'learning_rate/action_query_proj': 1.123166078736081e-05, 'learning_rate/base': 2.8079151968402024e-06, 'epoch': 0.71}) +[2026-06-30 08:06:07,462][starVLA.training.train_starvla][INFO] - Step 1647, Loss: {'train/loss': 0.07212721556425095, 'train/grad_norm_pre_clip': 2.5469555854797363, 'timing/data': 0.00020532705821096897, 'timing/model': 1.231240721186623, 'learning_rate/action_model': 1.1191912485658138e-05, 'learning_rate/action_query_proj': 1.1191912485658138e-05, 'learning_rate/base': 2.7979781214145344e-06, 'epoch': 0.71}) +[2026-06-30 08:06:40,134][starVLA.training.train_starvla][INFO] - Step 1648, Loss: {'train/loss': 0.5939562320709229, 'train/grad_norm_pre_clip': 2.00449538230896, 'timing/data': 0.0002715219743549824, 'timing/model': 1.2352904139552265, 'learning_rate/action_model': 1.1152264873854796e-05, 'learning_rate/action_query_proj': 1.1152264873854796e-05, 'learning_rate/base': 2.788066218463699e-06, 'epoch': 0.71}) +[2026-06-30 08:07:12,766][starVLA.training.train_starvla][INFO] - Step 1649, Loss: {'train/loss': 0.012433252297341824, 'train/grad_norm_pre_clip': 1.6272668838500977, 'timing/data': 0.0005285169463604689, 'timing/model': 1.2285933098755777, 'learning_rate/action_model': 1.1112718049777316e-05, 'learning_rate/action_query_proj': 1.1112718049777316e-05, 'learning_rate/base': 2.778179512444329e-06, 'epoch': 0.71}) +[2026-06-30 08:07:45,436][starVLA.training.train_starvla][INFO] - Step 1650, Loss: {'train/loss': 0.02193836309015751, 'train/grad_norm_pre_clip': 3.422447443008423, 'timing/data': 0.00019985600374639034, 'timing/model': 1.2308497680351138, 'learning_rate/action_model': 1.1073272111003576e-05, 'learning_rate/action_query_proj': 1.1073272111003576e-05, 'learning_rate/base': 2.768318027750894e-06, 'epoch': 0.71}) +[2026-06-30 08:08:18,102][starVLA.training.train_starvla][INFO] - Step 1651, Loss: {'train/loss': 0.17099332809448242, 'train/grad_norm_pre_clip': 3.0992627143859863, 'timing/data': 0.00042715505696833134, 'timing/model': 1.2305419989861548, 'learning_rate/action_model': 1.1033927154862495e-05, 'learning_rate/action_query_proj': 1.1033927154862495e-05, 'learning_rate/base': 2.7584817887156238e-06, 'epoch': 0.71}) +[2026-06-30 08:08:50,783][starVLA.training.train_starvla][INFO] - Step 1652, Loss: {'train/loss': 0.6603553891181946, 'train/grad_norm_pre_clip': 2.920067310333252, 'timing/data': 0.0002709908876568079, 'timing/model': 1.2312709151301533, 'learning_rate/action_model': 1.0994683278433836e-05, 'learning_rate/action_query_proj': 1.0994683278433836e-05, 'learning_rate/base': 2.748670819608459e-06, 'epoch': 0.71}) +[2026-06-30 08:09:23,481][starVLA.training.train_starvla][INFO] - Step 1653, Loss: {'train/loss': 0.24779193103313446, 'train/grad_norm_pre_clip': 2.039888858795166, 'timing/data': 0.00020182994194328785, 'timing/model': 1.2311157409567386, 'learning_rate/action_model': 1.0955540578547968e-05, 'learning_rate/action_query_proj': 1.0955540578547968e-05, 'learning_rate/base': 2.738885144636992e-06, 'epoch': 0.71}) +[2026-06-30 08:09:56,135][starVLA.training.train_starvla][INFO] - Step 1654, Loss: {'train/loss': 0.05853249877691269, 'train/grad_norm_pre_clip': 2.0509002208709717, 'timing/data': 0.00026716385036706924, 'timing/model': 1.2287264550104737, 'learning_rate/action_model': 1.0916499151785618e-05, 'learning_rate/action_query_proj': 1.0916499151785618e-05, 'learning_rate/base': 2.7291247879464044e-06, 'epoch': 0.71}) +[2026-06-30 08:10:28,816][starVLA.training.train_starvla][INFO] - Step 1655, Loss: {'train/loss': 0.055867038667201996, 'train/grad_norm_pre_clip': 1.9110594987869263, 'timing/data': 0.0001929940190166235, 'timing/model': 1.2327625029720366, 'learning_rate/action_model': 1.0877559094477621e-05, 'learning_rate/action_query_proj': 1.0877559094477621e-05, 'learning_rate/base': 2.7193897736194054e-06, 'epoch': 0.71}) +[2026-06-30 08:11:01,505][starVLA.training.train_starvla][INFO] - Step 1656, Loss: {'train/loss': 0.013052982278168201, 'train/grad_norm_pre_clip': 2.2387800216674805, 'timing/data': 0.00023750797845423222, 'timing/model': 1.2315537189133465, 'learning_rate/action_model': 1.08387205027047e-05, 'learning_rate/action_query_proj': 1.08387205027047e-05, 'learning_rate/base': 2.709680125676175e-06, 'epoch': 0.71}) +[2026-06-30 08:11:34,170][starVLA.training.train_starvla][INFO] - Step 1657, Loss: {'train/loss': 0.12223614752292633, 'train/grad_norm_pre_clip': 1.7855684757232666, 'timing/data': 0.00021303188987076283, 'timing/model': 1.2301017548888922, 'learning_rate/action_model': 1.0799983472297201e-05, 'learning_rate/action_query_proj': 1.0799983472297201e-05, 'learning_rate/base': 2.6999958680743003e-06, 'epoch': 0.71}) +[2026-06-30 08:12:06,860][starVLA.training.train_starvla][INFO] - Step 1658, Loss: {'train/loss': 0.3111285865306854, 'train/grad_norm_pre_clip': 2.4425220489501953, 'timing/data': 0.00018676207400858402, 'timing/model': 1.2304194781463593, 'learning_rate/action_model': 1.0761348098834938e-05, 'learning_rate/action_query_proj': 1.0761348098834938e-05, 'learning_rate/base': 2.6903370247087344e-06, 'epoch': 0.71}) +[2026-06-30 08:12:39,755][starVLA.training.train_starvla][INFO] - Step 1659, Loss: {'train/loss': 0.014379444532096386, 'train/grad_norm_pre_clip': 2.1513283252716064, 'timing/data': 0.00026039103977382183, 'timing/model': 1.2307326218578964, 'learning_rate/action_model': 1.0722814477646802e-05, 'learning_rate/action_query_proj': 1.0722814477646802e-05, 'learning_rate/base': 2.6807036194117005e-06, 'epoch': 0.71}) +[2026-06-30 08:13:12,416][starVLA.training.train_starvla][INFO] - Step 1660, Loss: {'train/loss': 0.0011671147076413035, 'train/grad_norm_pre_clip': 2.5299553871154785, 'timing/data': 0.0003082011826336384, 'timing/model': 1.2323367709759623, 'learning_rate/action_model': 1.0684382703810709e-05, 'learning_rate/action_query_proj': 1.0684382703810709e-05, 'learning_rate/base': 2.6710956759526772e-06, 'epoch': 0.71}) +[2026-06-30 08:13:45,087][starVLA.training.train_starvla][INFO] - Step 1661, Loss: {'train/loss': 0.0016896333545446396, 'train/grad_norm_pre_clip': 1.5278823375701904, 'timing/data': 0.00019632093608379364, 'timing/model': 1.23028424102813, 'learning_rate/action_model': 1.0646052872153222e-05, 'learning_rate/action_query_proj': 1.0646052872153222e-05, 'learning_rate/base': 2.6615132180383055e-06, 'epoch': 0.71}) +[2026-06-30 08:14:17,747][starVLA.training.train_starvla][INFO] - Step 1662, Loss: {'train/loss': 0.04467521980404854, 'train/grad_norm_pre_clip': 1.877824068069458, 'timing/data': 0.00037991697899997234, 'timing/model': 1.2308952738530934, 'learning_rate/action_model': 1.0607825077249419e-05, 'learning_rate/action_query_proj': 1.0607825077249419e-05, 'learning_rate/base': 2.6519562693123547e-06, 'epoch': 0.72}) +[2026-06-30 08:14:50,401][starVLA.training.train_starvla][INFO] - Step 1663, Loss: {'train/loss': 0.0016935884486883879, 'train/grad_norm_pre_clip': 0.7651724815368652, 'timing/data': 0.00020220200531184673, 'timing/model': 1.235582116059959, 'learning_rate/action_model': 1.056969941342254e-05, 'learning_rate/action_query_proj': 1.056969941342254e-05, 'learning_rate/base': 2.642424853355635e-06, 'epoch': 0.72}) +[2026-06-30 08:15:23,055][starVLA.training.train_starvla][INFO] - Step 1664, Loss: {'train/loss': 0.03408249095082283, 'train/grad_norm_pre_clip': 3.6709578037261963, 'timing/data': 0.00028947601094841957, 'timing/model': 1.2308163358829916, 'learning_rate/action_model': 1.0531675974743907e-05, 'learning_rate/action_query_proj': 1.0531675974743907e-05, 'learning_rate/base': 2.632918993685977e-06, 'epoch': 0.72}) +[2026-06-30 08:15:55,690][starVLA.training.train_starvla][INFO] - Step 1665, Loss: {'train/loss': 2.363259955018293e-05, 'train/grad_norm_pre_clip': 3.3615705966949463, 'timing/data': 0.00019865483045578003, 'timing/model': 1.2255145949311554, 'learning_rate/action_model': 1.0493754855032562e-05, 'learning_rate/action_query_proj': 1.0493754855032562e-05, 'learning_rate/base': 2.6234387137581405e-06, 'epoch': 0.72}) +[2026-06-30 08:16:28,329][starVLA.training.train_starvla][INFO] - Step 1666, Loss: {'train/loss': 0.11169734597206116, 'train/grad_norm_pre_clip': 2.030151844024658, 'timing/data': 0.0001983940601348877, 'timing/model': 1.2313571330159903, 'learning_rate/action_model': 1.04559361478551e-05, 'learning_rate/action_query_proj': 1.04559361478551e-05, 'learning_rate/base': 2.613984036963775e-06, 'epoch': 0.72}) +[2026-06-30 08:17:00,960][starVLA.training.train_starvla][INFO] - Step 1667, Loss: {'train/loss': 0.12838928401470184, 'train/grad_norm_pre_clip': 2.2244067192077637, 'timing/data': 0.0002890150062739849, 'timing/model': 1.2270151409320533, 'learning_rate/action_model': 1.0418219946525402e-05, 'learning_rate/action_query_proj': 1.0418219946525402e-05, 'learning_rate/base': 2.6045549866313504e-06, 'epoch': 0.72}) +[2026-06-30 08:17:33,625][starVLA.training.train_starvla][INFO] - Step 1668, Loss: {'train/loss': 0.23483431339263916, 'train/grad_norm_pre_clip': 3.1469802856445312, 'timing/data': 0.00034850696101784706, 'timing/model': 1.231151008978486, 'learning_rate/action_model': 1.0380606344104478e-05, 'learning_rate/action_query_proj': 1.0380606344104478e-05, 'learning_rate/base': 2.5951515860261196e-06, 'epoch': 0.72}) +[2026-06-30 08:18:06,282][starVLA.training.train_starvla][INFO] - Step 1669, Loss: {'train/loss': 0.00034826970659196377, 'train/grad_norm_pre_clip': 2.2823939323425293, 'timing/data': 0.0004563811235129833, 'timing/model': 1.2430015858262777, 'learning_rate/action_model': 1.0343095433400141e-05, 'learning_rate/action_query_proj': 1.0343095433400141e-05, 'learning_rate/base': 2.5857738583500352e-06, 'epoch': 0.72}) +[2026-06-30 08:18:38,984][starVLA.training.train_starvla][INFO] - Step 1670, Loss: {'train/loss': 0.411308616399765, 'train/grad_norm_pre_clip': 2.549119472503662, 'timing/data': 0.00021868199110031128, 'timing/model': 1.229894582880661, 'learning_rate/action_model': 1.0305687306966827e-05, 'learning_rate/action_query_proj': 1.0305687306966827e-05, 'learning_rate/base': 2.5764218267417067e-06, 'epoch': 0.72}) +[2026-06-30 08:19:11,623][starVLA.training.train_starvla][INFO] - Step 1671, Loss: {'train/loss': 0.036926135420799255, 'train/grad_norm_pre_clip': 2.136765718460083, 'timing/data': 0.00020756199955940247, 'timing/model': 1.2346819879021496, 'learning_rate/action_model': 1.0268382057105363e-05, 'learning_rate/action_query_proj': 1.0268382057105363e-05, 'learning_rate/base': 2.5670955142763407e-06, 'epoch': 0.72}) +[2026-06-30 08:19:44,291][starVLA.training.train_starvla][INFO] - Step 1672, Loss: {'train/loss': 0.04652835428714752, 'train/grad_norm_pre_clip': 0.9068621397018433, 'timing/data': 0.00019774306565523148, 'timing/model': 1.2301305199507624, 'learning_rate/action_model': 1.0231179775862768e-05, 'learning_rate/action_query_proj': 1.0231179775862768e-05, 'learning_rate/base': 2.557794943965692e-06, 'epoch': 0.72}) +[2026-06-30 08:20:16,952][starVLA.training.train_starvla][INFO] - Step 1673, Loss: {'train/loss': 0.0004943894455209374, 'train/grad_norm_pre_clip': 2.9946937561035156, 'timing/data': 0.00020222109742462635, 'timing/model': 1.230866672936827, 'learning_rate/action_model': 1.0194080555031959e-05, 'learning_rate/action_query_proj': 1.0194080555031959e-05, 'learning_rate/base': 2.5485201387579897e-06, 'epoch': 0.72}) +[2026-06-30 08:20:49,579][starVLA.training.train_starvla][INFO] - Step 1674, Loss: {'train/loss': 0.1457965075969696, 'train/grad_norm_pre_clip': 3.562544822692871, 'timing/data': 0.000194967957213521, 'timing/model': 1.2342903960961848, 'learning_rate/action_model': 1.015708448615158e-05, 'learning_rate/action_query_proj': 1.015708448615158e-05, 'learning_rate/base': 2.539271121537895e-06, 'epoch': 0.72}) +[2026-06-30 08:21:22,259][starVLA.training.train_starvla][INFO] - Step 1675, Loss: {'train/loss': 6.552832201123238e-05, 'train/grad_norm_pre_clip': 4.599233627319336, 'timing/data': 0.00034044194035232067, 'timing/model': 1.2318297049496323, 'learning_rate/action_model': 1.0120191660505735e-05, 'learning_rate/action_query_proj': 1.0120191660505735e-05, 'learning_rate/base': 2.530047915126434e-06, 'epoch': 0.72}) +[2026-06-30 08:21:54,935][starVLA.training.train_starvla][INFO] - Step 1676, Loss: {'train/loss': 0.0016726120375096798, 'train/grad_norm_pre_clip': 3.61554217338562, 'timing/data': 0.0005687039811164141, 'timing/model': 1.2307033620309085, 'learning_rate/action_model': 1.0083402169123843e-05, 'learning_rate/action_query_proj': 1.0083402169123843e-05, 'learning_rate/base': 2.5208505422809606e-06, 'epoch': 0.72}) +[2026-06-30 08:22:27,620][starVLA.training.train_starvla][INFO] - Step 1677, Loss: {'train/loss': 0.15237392485141754, 'train/grad_norm_pre_clip': 2.0394184589385986, 'timing/data': 0.00021455506794154644, 'timing/model': 1.2342450830619782, 'learning_rate/action_model': 1.004671610278027e-05, 'learning_rate/action_query_proj': 1.004671610278027e-05, 'learning_rate/base': 2.5116790256950677e-06, 'epoch': 0.72}) +[2026-06-30 08:23:00,304][starVLA.training.train_starvla][INFO] - Step 1678, Loss: {'train/loss': 0.5617563724517822, 'train/grad_norm_pre_clip': 3.245192050933838, 'timing/data': 0.0004054461605846882, 'timing/model': 1.2319592989515513, 'learning_rate/action_model': 1.001013355199427e-05, 'learning_rate/action_query_proj': 1.001013355199427e-05, 'learning_rate/base': 2.5025333879985674e-06, 'epoch': 0.72}) +[2026-06-30 08:23:32,932][starVLA.training.train_starvla][INFO] - Step 1679, Loss: {'train/loss': 0.012636088766157627, 'train/grad_norm_pre_clip': 2.697542905807495, 'timing/data': 0.00029908400028944016, 'timing/model': 1.230630812002346, 'learning_rate/action_model': 9.973654607029631e-06, 'learning_rate/action_query_proj': 9.973654607029631e-06, 'learning_rate/base': 2.4934136517574077e-06, 'epoch': 0.72}) +[2026-06-30 08:24:05,563][starVLA.training.train_starvla][INFO] - Step 1680, Loss: {'train/loss': 0.00020229900837875903, 'train/grad_norm_pre_clip': 2.7606356143951416, 'timing/data': 0.0002503229770809412, 'timing/model': 1.2340938788838685, 'learning_rate/action_model': 9.937279357894558e-06, 'learning_rate/action_query_proj': 9.937279357894558e-06, 'learning_rate/base': 2.4843198394736394e-06, 'epoch': 0.72}) +[2026-06-30 08:24:38,237][starVLA.training.train_starvla][INFO] - Step 1681, Loss: {'train/loss': 0.00040457857539877295, 'train/grad_norm_pre_clip': 2.07676100730896, 'timing/data': 0.00025032320991158485, 'timing/model': 1.2323939260095358, 'learning_rate/action_model': 9.901007894341311e-06, 'learning_rate/action_query_proj': 9.901007894341311e-06, 'learning_rate/base': 2.4752519735853277e-06, 'epoch': 0.72}) +[2026-06-30 08:25:10,886][starVLA.training.train_starvla][INFO] - Step 1682, Loss: {'train/loss': 0.0001377706357743591, 'train/grad_norm_pre_clip': 1.9786784648895264, 'timing/data': 0.0002570957876741886, 'timing/model': 1.2291200298350304, 'learning_rate/action_model': 9.864840305866162e-06, 'learning_rate/action_query_proj': 9.864840305866162e-06, 'learning_rate/base': 2.4662100764665406e-06, 'epoch': 0.72}) +[2026-06-30 08:25:43,461][starVLA.training.train_starvla][INFO] - Step 1683, Loss: {'train/loss': 0.00011745915981009603, 'train/grad_norm_pre_clip': 2.082327127456665, 'timing/data': 0.00019683199934661388, 'timing/model': 1.2282243790104985, 'learning_rate/action_model': 9.828776681709025e-06, 'learning_rate/action_query_proj': 9.828776681709025e-06, 'learning_rate/base': 2.4571941704272563e-06, 'epoch': 0.72}) +[2026-06-30 08:26:16,090][starVLA.training.train_starvla][INFO] - Step 1684, Loss: {'train/loss': 3.918743823305704e-05, 'train/grad_norm_pre_clip': 3.9665753841400146, 'timing/data': 0.0001905490644276142, 'timing/model': 1.2320847720839083, 'learning_rate/action_model': 9.792817110853306e-06, 'learning_rate/action_query_proj': 9.792817110853306e-06, 'learning_rate/base': 2.4482042777133265e-06, 'epoch': 0.72}) +[2026-06-30 08:26:48,740][starVLA.training.train_starvla][INFO] - Step 1685, Loss: {'train/loss': 2.2261645426624455e-05, 'train/grad_norm_pre_clip': 2.6710710525512695, 'timing/data': 0.00019362615421414375, 'timing/model': 1.231599068036303, 'learning_rate/action_model': 9.756961682025667e-06, 'learning_rate/action_query_proj': 9.756961682025667e-06, 'learning_rate/base': 2.439240420506417e-06, 'epoch': 0.73}) +[2026-06-30 08:27:21,384][starVLA.training.train_starvla][INFO] - Step 1686, Loss: {'train/loss': 0.21642029285430908, 'train/grad_norm_pre_clip': 1.4930607080459595, 'timing/data': 0.000197211978957057, 'timing/model': 1.2316887939814478, 'learning_rate/action_model': 9.721210483695828e-06, 'learning_rate/action_query_proj': 9.721210483695828e-06, 'learning_rate/base': 2.430302620923957e-06, 'epoch': 0.73}) +[2026-06-30 08:27:53,960][starVLA.training.train_starvla][INFO] - Step 1687, Loss: {'train/loss': 0.020302709192037582, 'train/grad_norm_pre_clip': 2.293276309967041, 'timing/data': 0.00026582181453704834, 'timing/model': 1.2354353878181428, 'learning_rate/action_model': 9.685563604076315e-06, 'learning_rate/action_query_proj': 9.685563604076315e-06, 'learning_rate/base': 2.4213909010190787e-06, 'epoch': 0.73}) +[2026-06-30 08:28:26,624][starVLA.training.train_starvla][INFO] - Step 1688, Loss: {'train/loss': 0.0006076542194932699, 'train/grad_norm_pre_clip': 2.0079410076141357, 'timing/data': 0.0002004790585488081, 'timing/model': 1.230836576083675, 'learning_rate/action_model': 9.650021131122247e-06, 'learning_rate/action_query_proj': 9.650021131122247e-06, 'learning_rate/base': 2.4125052827805617e-06, 'epoch': 0.73}) +[2026-06-30 08:28:59,313][starVLA.training.train_starvla][INFO] - Step 1689, Loss: {'train/loss': 0.0001209130568895489, 'train/grad_norm_pre_clip': 2.050111770629883, 'timing/data': 0.0001974829938262701, 'timing/model': 1.230842568911612, 'learning_rate/action_model': 9.614583152531145e-06, 'learning_rate/action_query_proj': 9.614583152531145e-06, 'learning_rate/base': 2.403645788132786e-06, 'epoch': 0.73}) +[2026-06-30 08:29:31,955][starVLA.training.train_starvla][INFO] - Step 1690, Loss: {'train/loss': 0.003103886963799596, 'train/grad_norm_pre_clip': 3.5670182704925537, 'timing/data': 0.00019701197743415833, 'timing/model': 1.2306499418336898, 'learning_rate/action_model': 9.579249755742715e-06, 'learning_rate/action_query_proj': 9.579249755742715e-06, 'learning_rate/base': 2.3948124389356787e-06, 'epoch': 0.73}) +[2026-06-30 08:30:04,657][starVLA.training.train_starvla][INFO] - Step 1691, Loss: {'train/loss': 0.00017309506074525416, 'train/grad_norm_pre_clip': 3.9721856117248535, 'timing/data': 0.00020874408073723316, 'timing/model': 1.2298698001541197, 'learning_rate/action_model': 9.544021027938594e-06, 'learning_rate/action_query_proj': 9.544021027938594e-06, 'learning_rate/base': 2.3860052569846484e-06, 'epoch': 0.73}) +[2026-06-30 08:30:37,338][starVLA.training.train_starvla][INFO] - Step 1692, Loss: {'train/loss': 0.018217019736766815, 'train/grad_norm_pre_clip': 2.874453544616699, 'timing/data': 0.00036948686465620995, 'timing/model': 1.2299978961236775, 'learning_rate/action_model': 9.508897056042153e-06, 'learning_rate/action_query_proj': 9.508897056042153e-06, 'learning_rate/base': 2.377224264010538e-06, 'epoch': 0.73}) +[2026-06-30 08:31:09,945][starVLA.training.train_starvla][INFO] - Step 1693, Loss: {'train/loss': 0.00039926491444930434, 'train/grad_norm_pre_clip': 2.057798385620117, 'timing/data': 0.00026826607063412666, 'timing/model': 1.2305696008261293, 'learning_rate/action_model': 9.473877926718305e-06, 'learning_rate/action_query_proj': 9.473877926718305e-06, 'learning_rate/base': 2.3684694816795762e-06, 'epoch': 0.73}) +[2026-06-30 08:31:42,611][starVLA.training.train_starvla][INFO] - Step 1694, Loss: {'train/loss': 1.0520130672375672e-05, 'train/grad_norm_pre_clip': 1.2480804920196533, 'timing/data': 0.0003433478996157646, 'timing/model': 1.2324202230665833, 'learning_rate/action_model': 9.438963726373286e-06, 'learning_rate/action_query_proj': 9.438963726373286e-06, 'learning_rate/base': 2.3597409315933216e-06, 'epoch': 0.73}) +[2026-06-30 08:32:15,297][starVLA.training.train_starvla][INFO] - Step 1695, Loss: {'train/loss': 1.576525937707629e-05, 'train/grad_norm_pre_clip': 1.8631539344787598, 'timing/data': 0.0002948860637843609, 'timing/model': 1.2349357139319181, 'learning_rate/action_model': 9.404154541154408e-06, 'learning_rate/action_query_proj': 9.404154541154408e-06, 'learning_rate/base': 2.351038635288602e-06, 'epoch': 0.73}) +[2026-06-30 08:32:48,009][starVLA.training.train_starvla][INFO] - Step 1696, Loss: {'train/loss': 2.44667571678292e-05, 'train/grad_norm_pre_clip': 1.531241774559021, 'timing/data': 0.00020842300727963448, 'timing/model': 1.2283544728998095, 'learning_rate/action_model': 9.369450456949864e-06, 'learning_rate/action_query_proj': 9.369450456949864e-06, 'learning_rate/base': 2.342362614237466e-06, 'epoch': 0.73}) +[2026-06-30 08:33:20,679][starVLA.training.train_starvla][INFO] - Step 1697, Loss: {'train/loss': 0.00621666107326746, 'train/grad_norm_pre_clip': 1.1956757307052612, 'timing/data': 0.00019823387265205383, 'timing/model': 1.227006747853011, 'learning_rate/action_model': 9.334851559388536e-06, 'learning_rate/action_query_proj': 9.334851559388536e-06, 'learning_rate/base': 2.333712889847134e-06, 'epoch': 0.73}) +[2026-06-30 08:33:53,337][starVLA.training.train_starvla][INFO] - Step 1698, Loss: {'train/loss': 0.04732997715473175, 'train/grad_norm_pre_clip': 1.5178617238998413, 'timing/data': 0.0002948169130831957, 'timing/model': 1.229833984049037, 'learning_rate/action_model': 9.300357933839786e-06, 'learning_rate/action_query_proj': 9.300357933839786e-06, 'learning_rate/base': 2.3250894834599466e-06, 'epoch': 0.73}) +[2026-06-30 08:34:26,044][starVLA.training.train_starvla][INFO] - Step 1699, Loss: {'train/loss': 1.0000542402267456, 'train/grad_norm_pre_clip': 3.469093084335327, 'timing/data': 0.00020611891523003578, 'timing/model': 1.230535953072831, 'learning_rate/action_model': 9.265969665413165e-06, 'learning_rate/action_query_proj': 9.265969665413165e-06, 'learning_rate/base': 2.3164924163532913e-06, 'epoch': 0.73}) +[2026-06-30 08:34:58,721][starVLA.training.train_starvla][INFO] - Step 1700, Loss: {'train/loss': 0.01664959266781807, 'train/grad_norm_pre_clip': 2.7512288093566895, 'timing/data': 0.0002121098805218935, 'timing/model': 1.231918518198654, 'learning_rate/action_model': 9.231686838958345e-06, 'learning_rate/action_query_proj': 9.231686838958345e-06, 'learning_rate/base': 2.3079217097395862e-06, 'epoch': 0.73}) +[2026-06-30 08:35:31,383][starVLA.training.train_starvla][INFO] - Step 1701, Loss: {'train/loss': 0.0031312431674450636, 'train/grad_norm_pre_clip': 2.8830881118774414, 'timing/data': 0.0003141830675303936, 'timing/model': 1.2284538699313998, 'learning_rate/action_model': 9.197509539064776e-06, 'learning_rate/action_query_proj': 9.197509539064776e-06, 'learning_rate/base': 2.299377384766194e-06, 'epoch': 0.73}) +[2026-06-30 08:36:04,062][starVLA.training.train_starvla][INFO] - Step 1702, Loss: {'train/loss': 7.09501764504239e-05, 'train/grad_norm_pre_clip': 3.323777675628662, 'timing/data': 0.00020722090266644955, 'timing/model': 1.2308214260265231, 'learning_rate/action_model': 9.163437850061546e-06, 'learning_rate/action_query_proj': 9.163437850061546e-06, 'learning_rate/base': 2.2908594625153866e-06, 'epoch': 0.73}) +[2026-06-30 08:36:36,725][starVLA.training.train_starvla][INFO] - Step 1703, Loss: {'train/loss': 0.0011786249233409762, 'train/grad_norm_pre_clip': 2.2315731048583984, 'timing/data': 0.00023940205574035645, 'timing/model': 1.2313392960932106, 'learning_rate/action_model': 9.129471856017157e-06, 'learning_rate/action_query_proj': 9.129471856017157e-06, 'learning_rate/base': 2.282367964004289e-06, 'epoch': 0.73}) +[2026-06-30 08:37:09,393][starVLA.training.train_starvla][INFO] - Step 1704, Loss: {'train/loss': 0.13469620048999786, 'train/grad_norm_pre_clip': 2.8884787559509277, 'timing/data': 0.000330423004925251, 'timing/model': 1.228904837043956, 'learning_rate/action_model': 9.095611640739338e-06, 'learning_rate/action_query_proj': 9.095611640739338e-06, 'learning_rate/base': 2.2739029101848346e-06, 'epoch': 0.73}) +[2026-06-30 08:37:42,034][starVLA.training.train_starvla][INFO] - Step 1705, Loss: {'train/loss': 0.014586778357625008, 'train/grad_norm_pre_clip': 2.716298818588257, 'timing/data': 0.0002024020068347454, 'timing/model': 1.227428657002747, 'learning_rate/action_model': 9.0618572877748e-06, 'learning_rate/action_query_proj': 9.0618572877748e-06, 'learning_rate/base': 2.2654643219437e-06, 'epoch': 0.73}) +[2026-06-30 08:38:14,690][starVLA.training.train_starvla][INFO] - Step 1706, Loss: {'train/loss': 0.000885210873093456, 'train/grad_norm_pre_clip': 3.3145639896392822, 'timing/data': 0.00020781205967068672, 'timing/model': 1.2285420140251517, 'learning_rate/action_model': 9.028208880409045e-06, 'learning_rate/action_query_proj': 9.028208880409045e-06, 'learning_rate/base': 2.2570522201022613e-06, 'epoch': 0.73}) +[2026-06-30 08:38:47,330][starVLA.training.train_starvla][INFO] - Step 1707, Loss: {'train/loss': 0.006601642817258835, 'train/grad_norm_pre_clip': 2.3488643169403076, 'timing/data': 0.0001974431797862053, 'timing/model': 1.2294224039651453, 'learning_rate/action_model': 8.994666501666169e-06, 'learning_rate/action_query_proj': 8.994666501666169e-06, 'learning_rate/base': 2.248666625416542e-06, 'epoch': 0.73}) +[2026-06-30 08:39:20,008][starVLA.training.train_starvla][INFO] - Step 1708, Loss: {'train/loss': 0.04135829955339432, 'train/grad_norm_pre_clip': 1.8359043598175049, 'timing/data': 0.0004131689202040434, 'timing/model': 1.2334780448582023, 'learning_rate/action_model': 8.961230234308678e-06, 'learning_rate/action_query_proj': 8.961230234308678e-06, 'learning_rate/base': 2.2403075585771696e-06, 'epoch': 0.73}) +[2026-06-30 08:39:52,676][starVLA.training.train_starvla][INFO] - Step 1709, Loss: {'train/loss': 0.02479095384478569, 'train/grad_norm_pre_clip': 2.8173208236694336, 'timing/data': 0.0002860401291400194, 'timing/model': 1.2304510860703886, 'learning_rate/action_model': 8.927900160837225e-06, 'learning_rate/action_query_proj': 8.927900160837225e-06, 'learning_rate/base': 2.231975040209306e-06, 'epoch': 0.74}) +[2026-06-30 08:40:25,368][starVLA.training.train_starvla][INFO] - Step 1710, Loss: {'train/loss': 0.001480672974139452, 'train/grad_norm_pre_clip': 1.0686421394348145, 'timing/data': 0.0002042059786617756, 'timing/model': 1.2342902701348066, 'learning_rate/action_model': 8.894676363490453e-06, 'learning_rate/action_query_proj': 8.894676363490453e-06, 'learning_rate/base': 2.2236690908726132e-06, 'epoch': 0.74}) +[2026-06-30 08:40:58,032][starVLA.training.train_starvla][INFO] - Step 1711, Loss: {'train/loss': 2.154632966266945e-05, 'train/grad_norm_pre_clip': 2.7917027473449707, 'timing/data': 0.00021018600091338158, 'timing/model': 1.2386464709416032, 'learning_rate/action_model': 8.861558924244773e-06, 'learning_rate/action_query_proj': 8.861558924244773e-06, 'learning_rate/base': 2.215389731061193e-06, 'epoch': 0.74}) +[2026-06-30 08:41:30,893][starVLA.training.train_starvla][INFO] - Step 1712, Loss: {'train/loss': 0.01381372008472681, 'train/grad_norm_pre_clip': 1.6176986694335938, 'timing/data': 0.0002562329173088074, 'timing/model': 1.2305908789858222, 'learning_rate/action_model': 8.828547924814188e-06, 'learning_rate/action_query_proj': 8.828547924814188e-06, 'learning_rate/base': 2.207136981203547e-06, 'epoch': 0.74}) +[2026-06-30 08:42:03,555][starVLA.training.train_starvla][INFO] - Step 1713, Loss: {'train/loss': 0.0065748803317546844, 'train/grad_norm_pre_clip': 1.6098912954330444, 'timing/data': 0.00020498689264059067, 'timing/model': 1.2284809739794582, 'learning_rate/action_model': 8.795643446650055e-06, 'learning_rate/action_query_proj': 8.795643446650055e-06, 'learning_rate/base': 2.1989108616625137e-06, 'epoch': 0.74}) +[2026-06-30 08:42:36,205][starVLA.training.train_starvla][INFO] - Step 1714, Loss: {'train/loss': 0.0001220450212713331, 'train/grad_norm_pre_clip': 2.0343587398529053, 'timing/data': 0.00033430103212594986, 'timing/model': 1.2299369368702173, 'learning_rate/action_model': 8.762845570940891e-06, 'learning_rate/action_query_proj': 8.762845570940891e-06, 'learning_rate/base': 2.1907113927352228e-06, 'epoch': 0.74}) +[2026-06-30 08:43:08,893][starVLA.training.train_starvla][INFO] - Step 1715, Loss: {'train/loss': 0.10828743875026703, 'train/grad_norm_pre_clip': 2.52112078666687, 'timing/data': 0.0002616939600557089, 'timing/model': 1.2298408029600978, 'learning_rate/action_model': 8.730154378612202e-06, 'learning_rate/action_query_proj': 8.730154378612202e-06, 'learning_rate/base': 2.1825385946530504e-06, 'epoch': 0.74}) +[2026-06-30 08:43:41,562][starVLA.training.train_starvla][INFO] - Step 1716, Loss: {'train/loss': 0.17649008333683014, 'train/grad_norm_pre_clip': 2.8677051067352295, 'timing/data': 0.00025391881354153156, 'timing/model': 1.2374851510394365, 'learning_rate/action_model': 8.697569950326267e-06, 'learning_rate/action_query_proj': 8.697569950326267e-06, 'learning_rate/base': 2.1743924875815667e-06, 'epoch': 0.74}) +[2026-06-30 08:44:14,220][starVLA.training.train_starvla][INFO] - Step 1717, Loss: {'train/loss': 8.01675832917681e-06, 'train/grad_norm_pre_clip': 3.1368377208709717, 'timing/data': 0.00046354392543435097, 'timing/model': 1.228944978211075, 'learning_rate/action_model': 8.665092366481902e-06, 'learning_rate/action_query_proj': 8.665092366481902e-06, 'learning_rate/base': 2.1662730916204756e-06, 'epoch': 0.74}) +[2026-06-30 08:44:46,898][starVLA.training.train_starvla][INFO] - Step 1718, Loss: {'train/loss': 0.03123389557003975, 'train/grad_norm_pre_clip': 3.5038201808929443, 'timing/data': 0.00026414915919303894, 'timing/model': 1.2325459609273821, 'learning_rate/action_model': 8.632721707214329e-06, 'learning_rate/action_query_proj': 8.632721707214329e-06, 'learning_rate/base': 2.158180426803582e-06, 'epoch': 0.74}) +[2026-06-30 08:45:19,583][starVLA.training.train_starvla][INFO] - Step 1719, Loss: {'train/loss': 1.642089046072215e-05, 'train/grad_norm_pre_clip': 1.9450173377990723, 'timing/data': 0.00025554210878908634, 'timing/model': 1.2288590259850025, 'learning_rate/action_model': 8.60045805239493e-06, 'learning_rate/action_query_proj': 8.60045805239493e-06, 'learning_rate/base': 2.1501145130987327e-06, 'epoch': 0.74}) +[2026-06-30 08:45:52,258][starVLA.training.train_starvla][INFO] - Step 1720, Loss: {'train/loss': 0.0010108822025358677, 'train/grad_norm_pre_clip': 2.603745460510254, 'timing/data': 0.00020675989799201488, 'timing/model': 1.2328355780337006, 'learning_rate/action_model': 8.56830148163107e-06, 'learning_rate/action_query_proj': 8.56830148163107e-06, 'learning_rate/base': 2.1420753704077676e-06, 'epoch': 0.74}) +[2026-06-30 08:46:24,928][starVLA.training.train_starvla][INFO] - Step 1721, Loss: {'train/loss': 3.0695151508552954e-05, 'train/grad_norm_pre_clip': 2.2564914226531982, 'timing/data': 0.0003873109817504883, 'timing/model': 1.229170415084809, 'learning_rate/action_model': 8.536252074265872e-06, 'learning_rate/action_query_proj': 8.536252074265872e-06, 'learning_rate/base': 2.134063018566468e-06, 'epoch': 0.74}) +[2026-06-30 08:46:57,567][starVLA.training.train_starvla][INFO] - Step 1722, Loss: {'train/loss': 0.00014012688188813627, 'train/grad_norm_pre_clip': 1.6779536008834839, 'timing/data': 0.00019072997383773327, 'timing/model': 1.2349920189008117, 'learning_rate/action_model': 8.50430990937809e-06, 'learning_rate/action_query_proj': 8.50430990937809e-06, 'learning_rate/base': 2.1260774773445224e-06, 'epoch': 0.74}) +[2026-06-30 08:47:30,229][starVLA.training.train_starvla][INFO] - Step 1723, Loss: {'train/loss': 0.26080086827278137, 'train/grad_norm_pre_clip': 1.7776755094528198, 'timing/data': 0.0001958890352398157, 'timing/model': 1.2290981630794704, 'learning_rate/action_model': 8.472475065781824e-06, 'learning_rate/action_query_proj': 8.472475065781824e-06, 'learning_rate/base': 2.118118766445456e-06, 'epoch': 0.74}) +[2026-06-30 08:48:02,883][starVLA.training.train_starvla][INFO] - Step 1724, Loss: {'train/loss': 0.007833157666027546, 'train/grad_norm_pre_clip': 3.229457139968872, 'timing/data': 0.0002943449653685093, 'timing/model': 1.2336603018920869, 'learning_rate/action_model': 8.440747622026389e-06, 'learning_rate/action_query_proj': 8.440747622026389e-06, 'learning_rate/base': 2.1101869055065972e-06, 'epoch': 0.74}) +[2026-06-30 08:48:35,532][starVLA.training.train_starvla][INFO] - Step 1725, Loss: {'train/loss': 0.0019477502210065722, 'train/grad_norm_pre_clip': 2.534428119659424, 'timing/data': 0.00020733103156089783, 'timing/model': 1.2256184651050717, 'learning_rate/action_model': 8.409127656396095e-06, 'learning_rate/action_query_proj': 8.409127656396095e-06, 'learning_rate/base': 2.1022819140990237e-06, 'epoch': 0.74}) +[2026-06-30 08:49:08,122][starVLA.training.train_starvla][INFO] - Step 1726, Loss: {'train/loss': 0.008908769115805626, 'train/grad_norm_pre_clip': 2.016951084136963, 'timing/data': 0.0001992459874600172, 'timing/model': 1.2275814411696047, 'learning_rate/action_model': 8.377615246910072e-06, 'learning_rate/action_query_proj': 8.377615246910072e-06, 'learning_rate/base': 2.094403811727518e-06, 'epoch': 0.74}) +[2026-06-30 08:49:40,784][starVLA.training.train_starvla][INFO] - Step 1727, Loss: {'train/loss': 0.0685499906539917, 'train/grad_norm_pre_clip': 2.1775715351104736, 'timing/data': 0.00024577416479587555, 'timing/model': 1.2353033919353038, 'learning_rate/action_model': 8.34621047132206e-06, 'learning_rate/action_query_proj': 8.34621047132206e-06, 'learning_rate/base': 2.086552617830515e-06, 'epoch': 0.74}) +[2026-06-30 08:50:13,437][starVLA.training.train_starvla][INFO] - Step 1728, Loss: {'train/loss': 0.0022291571367532015, 'train/grad_norm_pre_clip': 2.4398341178894043, 'timing/data': 0.0002843770198523998, 'timing/model': 1.2329575109761208, 'learning_rate/action_model': 8.314913407120209e-06, 'learning_rate/action_query_proj': 8.314913407120209e-06, 'learning_rate/base': 2.0787283517800523e-06, 'epoch': 0.74}) +[2026-06-30 08:50:46,130][starVLA.training.train_starvla][INFO] - Step 1729, Loss: {'train/loss': 0.01104724407196045, 'train/grad_norm_pre_clip': 3.6857364177703857, 'timing/data': 0.00019158096984028816, 'timing/model': 1.2294428350869566, 'learning_rate/action_model': 8.283724131526912e-06, 'learning_rate/action_query_proj': 8.283724131526912e-06, 'learning_rate/base': 2.070931032881728e-06, 'epoch': 0.74}) +[2026-06-30 08:51:18,772][starVLA.training.train_starvla][INFO] - Step 1730, Loss: {'train/loss': 0.3254327178001404, 'train/grad_norm_pre_clip': 2.3155763149261475, 'timing/data': 0.00038049789145588875, 'timing/model': 1.2333901559468359, 'learning_rate/action_model': 8.252642721498625e-06, 'learning_rate/action_query_proj': 8.252642721498625e-06, 'learning_rate/base': 2.0631606803746563e-06, 'epoch': 0.74}) +[2026-06-30 08:51:51,413][starVLA.training.train_starvla][INFO] - Step 1731, Loss: {'train/loss': 0.0037969835102558136, 'train/grad_norm_pre_clip': 1.80470609664917, 'timing/data': 0.00028743199072778225, 'timing/model': 1.2312969039194286, 'learning_rate/action_model': 8.221669253725627e-06, 'learning_rate/action_query_proj': 8.221669253725627e-06, 'learning_rate/base': 2.055417313431407e-06, 'epoch': 0.74}) +[2026-06-30 08:52:24,095][starVLA.training.train_starvla][INFO] - Step 1732, Loss: {'train/loss': 0.004096535034477711, 'train/grad_norm_pre_clip': 1.6923326253890991, 'timing/data': 0.0002565039321780205, 'timing/model': 1.231380868004635, 'learning_rate/action_model': 8.190803804631866e-06, 'learning_rate/action_query_proj': 8.190803804631866e-06, 'learning_rate/base': 2.0477009511579666e-06, 'epoch': 0.75}) +[2026-06-30 08:52:56,742][starVLA.training.train_starvla][INFO] - Step 1733, Loss: {'train/loss': 0.0007813215488567948, 'train/grad_norm_pre_clip': 1.7795941829681396, 'timing/data': 0.0002095550298690796, 'timing/model': 1.2285291131120175, 'learning_rate/action_model': 8.160046450374768e-06, 'learning_rate/action_query_proj': 8.160046450374768e-06, 'learning_rate/base': 2.040011612593692e-06, 'epoch': 0.75}) +[2026-06-30 08:53:29,375][starVLA.training.train_starvla][INFO] - Step 1734, Loss: {'train/loss': 0.14282381534576416, 'train/grad_norm_pre_clip': 2.870262861251831, 'timing/data': 0.00019469717517495155, 'timing/model': 1.2312535899691284, 'learning_rate/action_model': 8.129397266845067e-06, 'learning_rate/action_query_proj': 8.129397266845067e-06, 'learning_rate/base': 2.032349316711267e-06, 'epoch': 0.75}) +[2026-06-30 08:54:02,032][starVLA.training.train_starvla][INFO] - Step 1735, Loss: {'train/loss': 0.00024947349447757006, 'train/grad_norm_pre_clip': 3.086432933807373, 'timing/data': 0.0004226970486342907, 'timing/model': 1.230362334055826, 'learning_rate/action_model': 8.098856329666545e-06, 'learning_rate/action_query_proj': 8.098856329666545e-06, 'learning_rate/base': 2.0247140824166364e-06, 'epoch': 0.75}) +[2026-06-30 08:54:34,740][starVLA.training.train_starvla][INFO] - Step 1736, Loss: {'train/loss': 2.3185863028629683e-05, 'train/grad_norm_pre_clip': 1.9250214099884033, 'timing/data': 0.0003314560744911432, 'timing/model': 1.2307868220377713, 'learning_rate/action_model': 8.068423714195958e-06, 'learning_rate/action_query_proj': 8.068423714195958e-06, 'learning_rate/base': 2.0171059285489896e-06, 'epoch': 0.75}) +[2026-06-30 08:55:07,420][starVLA.training.train_starvla][INFO] - Step 1737, Loss: {'train/loss': 6.496883997897385e-06, 'train/grad_norm_pre_clip': 2.964540958404541, 'timing/data': 0.00025428994558751583, 'timing/model': 1.2359713991172612, 'learning_rate/action_model': 8.038099495522744e-06, 'learning_rate/action_query_proj': 8.038099495522744e-06, 'learning_rate/base': 2.009524873880686e-06, 'epoch': 0.75}) +[2026-06-30 08:55:40,102][starVLA.training.train_starvla][INFO] - Step 1738, Loss: {'train/loss': 0.14447148144245148, 'train/grad_norm_pre_clip': 3.107719659805298, 'timing/data': 0.0003486168570816517, 'timing/model': 1.2291598501615226, 'learning_rate/action_model': 8.0078837484689e-06, 'learning_rate/action_query_proj': 8.0078837484689e-06, 'learning_rate/base': 2.001970937117225e-06, 'epoch': 0.75}) +[2026-06-30 08:56:12,767][starVLA.training.train_starvla][INFO] - Step 1739, Loss: {'train/loss': 0.0802680253982544, 'train/grad_norm_pre_clip': 2.1548702716827393, 'timing/data': 0.00019973702728748322, 'timing/model': 1.231291436124593, 'learning_rate/action_model': 7.977776547588777e-06, 'learning_rate/action_query_proj': 7.977776547588777e-06, 'learning_rate/base': 1.994444136897194e-06, 'epoch': 0.75}) +[2026-06-30 08:56:45,387][starVLA.training.train_starvla][INFO] - Step 1740, Loss: {'train/loss': 3.099436071352102e-06, 'train/grad_norm_pre_clip': 1.90727961063385, 'timing/data': 0.0001980639062821865, 'timing/model': 1.2293511601164937, 'learning_rate/action_model': 7.947777967168903e-06, 'learning_rate/action_query_proj': 7.947777967168903e-06, 'learning_rate/base': 1.986944491792226e-06, 'epoch': 0.75}) +[2026-06-30 08:57:18,022][starVLA.training.train_starvla][INFO] - Step 1741, Loss: {'train/loss': 0.0014095973456278443, 'train/grad_norm_pre_clip': 1.0685136318206787, 'timing/data': 0.000202312134206295, 'timing/model': 1.232778947101906, 'learning_rate/action_model': 7.917888081227801e-06, 'learning_rate/action_query_proj': 7.917888081227801e-06, 'learning_rate/base': 1.9794720203069503e-06, 'epoch': 0.75}) +[2026-06-30 08:57:50,679][starVLA.training.train_starvla][INFO] - Step 1742, Loss: {'train/loss': 0.19095256924629211, 'train/grad_norm_pre_clip': 4.251046657562256, 'timing/data': 0.00026884698309004307, 'timing/model': 1.2305430811829865, 'learning_rate/action_model': 7.888106963515781e-06, 'learning_rate/action_query_proj': 7.888106963515781e-06, 'learning_rate/base': 1.9720267408789453e-06, 'epoch': 0.75}) +[2026-06-30 08:58:23,381][starVLA.training.train_starvla][INFO] - Step 1743, Loss: {'train/loss': 1.117571082431823e-05, 'train/grad_norm_pre_clip': 3.561297655105591, 'timing/data': 0.00029574683867394924, 'timing/model': 1.2300478969700634, 'learning_rate/action_model': 7.858434687514782e-06, 'learning_rate/action_query_proj': 7.858434687514782e-06, 'learning_rate/base': 1.9646086718786955e-06, 'epoch': 0.75}) +[2026-06-30 08:58:56,033][starVLA.training.train_starvla][INFO] - Step 1744, Loss: {'train/loss': 0.04137783125042915, 'train/grad_norm_pre_clip': 3.062135696411133, 'timing/data': 0.0002045158762484789, 'timing/model': 1.2280016071163118, 'learning_rate/action_model': 7.82887132643822e-06, 'learning_rate/action_query_proj': 7.82887132643822e-06, 'learning_rate/base': 1.957217831609555e-06, 'epoch': 0.75}) +[2026-06-30 08:59:28,613][starVLA.training.train_starvla][INFO] - Step 1745, Loss: {'train/loss': 0.025667309761047363, 'train/grad_norm_pre_clip': 5.089725494384766, 'timing/data': 0.0003232387825846672, 'timing/model': 1.231714135967195, 'learning_rate/action_model': 7.799416953230742e-06, 'learning_rate/action_query_proj': 7.799416953230742e-06, 'learning_rate/base': 1.9498542383076855e-06, 'epoch': 0.75}) +[2026-06-30 09:00:01,304][starVLA.training.train_starvla][INFO] - Step 1746, Loss: {'train/loss': 0.00039147076313383877, 'train/grad_norm_pre_clip': 1.740932583808899, 'timing/data': 0.00025650416500866413, 'timing/model': 1.2323673539794981, 'learning_rate/action_model': 7.770071640568078e-06, 'learning_rate/action_query_proj': 7.770071640568078e-06, 'learning_rate/base': 1.9425179101420195e-06, 'epoch': 0.75}) +[2026-06-30 09:00:33,964][starVLA.training.train_starvla][INFO] - Step 1747, Loss: {'train/loss': 1.2427433830453083e-05, 'train/grad_norm_pre_clip': 3.1377921104431152, 'timing/data': 0.00020429491996765137, 'timing/model': 1.2311343939509243, 'learning_rate/action_model': 7.740835460856864e-06, 'learning_rate/action_query_proj': 7.740835460856864e-06, 'learning_rate/base': 1.935208865214216e-06, 'epoch': 0.75}) +[2026-06-30 09:01:06,652][starVLA.training.train_starvla][INFO] - Step 1748, Loss: {'train/loss': 1.2039993634971324e-05, 'train/grad_norm_pre_clip': 2.841268539428711, 'timing/data': 0.00019057001918554306, 'timing/model': 1.2319458189886063, 'learning_rate/action_model': 7.711708486234484e-06, 'learning_rate/action_query_proj': 7.711708486234484e-06, 'learning_rate/base': 1.927927121558621e-06, 'epoch': 0.75}) +[2026-06-30 09:01:39,290][starVLA.training.train_starvla][INFO] - Step 1749, Loss: {'train/loss': 0.02463545650243759, 'train/grad_norm_pre_clip': 2.419528007507324, 'timing/data': 0.00024698604829609394, 'timing/model': 1.2336394670419395, 'learning_rate/action_model': 7.68269078856885e-06, 'learning_rate/action_query_proj': 7.68269078856885e-06, 'learning_rate/base': 1.9206726971422126e-06, 'epoch': 0.75}) +[2026-06-30 09:02:11,963][starVLA.training.train_starvla][INFO] - Step 1750, Loss: {'train/loss': 0.056033212691545486, 'train/grad_norm_pre_clip': 3.0615603923797607, 'timing/data': 0.00025282707065343857, 'timing/model': 1.2359656780026853, 'learning_rate/action_model': 7.653782439458235e-06, 'learning_rate/action_query_proj': 7.653782439458235e-06, 'learning_rate/base': 1.913445609864559e-06, 'epoch': 0.75}) +[2026-06-30 09:02:44,651][starVLA.training.train_starvla][INFO] - Step 1751, Loss: {'train/loss': 0.11750099062919617, 'train/grad_norm_pre_clip': 2.3957250118255615, 'timing/data': 0.0002443308476358652, 'timing/model': 1.2290602549910545, 'learning_rate/action_model': 7.624983510231129e-06, 'learning_rate/action_query_proj': 7.624983510231129e-06, 'learning_rate/base': 1.9062458775577824e-06, 'epoch': 0.75}) +[2026-06-30 09:03:17,276][starVLA.training.train_starvla][INFO] - Step 1752, Loss: {'train/loss': 1.4602934243157506e-05, 'train/grad_norm_pre_clip': 3.425441026687622, 'timing/data': 0.00020039803348481655, 'timing/model': 1.2291723270900548, 'learning_rate/action_model': 7.5962940719460195e-06, 'learning_rate/action_query_proj': 7.5962940719460195e-06, 'learning_rate/base': 1.8990735179865049e-06, 'epoch': 0.75}) +[2026-06-30 09:03:49,937][starVLA.training.train_starvla][INFO] - Step 1753, Loss: {'train/loss': 5.930633506068261e-06, 'train/grad_norm_pre_clip': 1.875549077987671, 'timing/data': 0.00020466581918299198, 'timing/model': 1.2324010678566992, 'learning_rate/action_model': 7.567714195391243e-06, 'learning_rate/action_query_proj': 7.567714195391243e-06, 'learning_rate/base': 1.8919285488478108e-06, 'epoch': 0.75}) +[2026-06-30 09:04:22,573][starVLA.training.train_starvla][INFO] - Step 1754, Loss: {'train/loss': 0.03075891174376011, 'train/grad_norm_pre_clip': 3.0850279331207275, 'timing/data': 0.000245522940531373, 'timing/model': 1.2352901170961559, 'learning_rate/action_model': 7.539243951084803e-06, 'learning_rate/action_query_proj': 7.539243951084803e-06, 'learning_rate/base': 1.8848109877712007e-06, 'epoch': 0.75}) +[2026-06-30 09:04:55,211][starVLA.training.train_starvla][INFO] - Step 1755, Loss: {'train/loss': 0.0297564510256052, 'train/grad_norm_pre_clip': 2.419738531112671, 'timing/data': 0.00019669090397655964, 'timing/model': 1.2293181018903852, 'learning_rate/action_model': 7.510883409274201e-06, 'learning_rate/action_query_proj': 7.510883409274201e-06, 'learning_rate/base': 1.8777208523185503e-06, 'epoch': 0.76}) +[2026-06-30 09:05:27,830][starVLA.training.train_starvla][INFO] - Step 1756, Loss: {'train/loss': 0.007684390526264906, 'train/grad_norm_pre_clip': 1.8961424827575684, 'timing/data': 0.00019729183986783028, 'timing/model': 1.2289299180265516, 'learning_rate/action_model': 7.482632639936259e-06, 'learning_rate/action_query_proj': 7.482632639936259e-06, 'learning_rate/base': 1.8706581599840648e-06, 'epoch': 0.76}) +[2026-06-30 09:06:00,473][starVLA.training.train_starvla][INFO] - Step 1757, Loss: {'train/loss': 0.024455171078443527, 'train/grad_norm_pre_clip': 6.352003574371338, 'timing/data': 0.0003785139415413141, 'timing/model': 1.2299293419346213, 'learning_rate/action_model': 7.454491712776938e-06, 'learning_rate/action_query_proj': 7.454491712776938e-06, 'learning_rate/base': 1.8636229281942346e-06, 'epoch': 0.76}) +[2026-06-30 09:06:33,160][starVLA.training.train_starvla][INFO] - Step 1758, Loss: {'train/loss': 0.00017211350495927036, 'train/grad_norm_pre_clip': 2.749969482421875, 'timing/data': 0.0002594690304249525, 'timing/model': 1.2338467580266297, 'learning_rate/action_model': 7.426460697231166e-06, 'learning_rate/action_query_proj': 7.426460697231166e-06, 'learning_rate/base': 1.8566151743077916e-06, 'epoch': 0.76}) +[2026-06-30 09:07:05,795][starVLA.training.train_starvla][INFO] - Step 1759, Loss: {'train/loss': 0.0009203501977026463, 'train/grad_norm_pre_clip': 1.6016249656677246, 'timing/data': 0.000209504971280694, 'timing/model': 1.2313028669450432, 'learning_rate/action_model': 7.398539662462714e-06, 'learning_rate/action_query_proj': 7.398539662462714e-06, 'learning_rate/base': 1.8496349156156786e-06, 'epoch': 0.76}) +[2026-06-30 09:07:38,472][starVLA.training.train_starvla][INFO] - Step 1760, Loss: {'train/loss': 0.021992981433868408, 'train/grad_norm_pre_clip': 1.1619864702224731, 'timing/data': 0.0002057680394500494, 'timing/model': 1.2301644459366798, 'learning_rate/action_model': 7.370728677363935e-06, 'learning_rate/action_query_proj': 7.370728677363935e-06, 'learning_rate/base': 1.8426821693409837e-06, 'epoch': 0.76}) +[2026-06-30 09:08:11,109][starVLA.training.train_starvla][INFO] - Step 1761, Loss: {'train/loss': 0.027371302247047424, 'train/grad_norm_pre_clip': 2.5706050395965576, 'timing/data': 0.00032739690504968166, 'timing/model': 1.2295354469679296, 'learning_rate/action_model': 7.343027810555686e-06, 'learning_rate/action_query_proj': 7.343027810555686e-06, 'learning_rate/base': 1.8357569526389216e-06, 'epoch': 0.76}) +[2026-06-30 09:08:43,776][starVLA.training.train_starvla][INFO] - Step 1762, Loss: {'train/loss': 0.7700052261352539, 'train/grad_norm_pre_clip': 2.483060836791992, 'timing/data': 0.00029460503719747066, 'timing/model': 1.2328503949102014, 'learning_rate/action_model': 7.315437130387083e-06, 'learning_rate/action_query_proj': 7.315437130387083e-06, 'learning_rate/base': 1.8288592825967708e-06, 'epoch': 0.76}) +[2026-06-30 09:09:16,467][starVLA.training.train_starvla][INFO] - Step 1763, Loss: {'train/loss': 0.010976698249578476, 'train/grad_norm_pre_clip': 2.684549331665039, 'timing/data': 0.0002029729075729847, 'timing/model': 1.229212247999385, 'learning_rate/action_model': 7.2879567049354115e-06, 'learning_rate/action_query_proj': 7.2879567049354115e-06, 'learning_rate/base': 1.8219891762338529e-06, 'epoch': 0.76}) +[2026-06-30 09:09:49,296][starVLA.training.train_starvla][INFO] - Step 1764, Loss: {'train/loss': 0.14604175090789795, 'train/grad_norm_pre_clip': 2.4055638313293457, 'timing/data': 0.0001958790235221386, 'timing/model': 1.2299093781039119, 'learning_rate/action_model': 7.260586602005874e-06, 'learning_rate/action_query_proj': 7.260586602005874e-06, 'learning_rate/base': 1.8151466505014685e-06, 'epoch': 0.76}) +[2026-06-30 09:10:21,998][starVLA.training.train_starvla][INFO] - Step 1765, Loss: {'train/loss': 0.006732994690537453, 'train/grad_norm_pre_clip': 3.1318342685699463, 'timing/data': 0.0002325188834220171, 'timing/model': 1.2357934322208166, 'learning_rate/action_model': 7.233326889131488e-06, 'learning_rate/action_query_proj': 7.233326889131488e-06, 'learning_rate/base': 1.808331722282872e-06, 'epoch': 0.76}) +[2026-06-30 09:10:54,668][starVLA.training.train_starvla][INFO] - Step 1766, Loss: {'train/loss': 0.0005125336465425789, 'train/grad_norm_pre_clip': 3.89668869972229, 'timing/data': 0.0002047359012067318, 'timing/model': 1.2326471721753478, 'learning_rate/action_model': 7.206177633572865e-06, 'learning_rate/action_query_proj': 7.206177633572865e-06, 'learning_rate/base': 1.8015444083932162e-06, 'epoch': 0.76}) +[2026-06-30 09:11:27,308][starVLA.training.train_starvla][INFO] - Step 1767, Loss: {'train/loss': 0.0011255182325839996, 'train/grad_norm_pre_clip': 1.3571140766143799, 'timing/data': 0.00018843612633645535, 'timing/model': 1.2292117360047996, 'learning_rate/action_model': 7.179138902318128e-06, 'learning_rate/action_query_proj': 7.179138902318128e-06, 'learning_rate/base': 1.794784725579532e-06, 'epoch': 0.76}) +[2026-06-30 09:11:59,971][starVLA.training.train_starvla][INFO] - Step 1768, Loss: {'train/loss': 0.05683395266532898, 'train/grad_norm_pre_clip': 1.7412607669830322, 'timing/data': 0.00019568903371691704, 'timing/model': 1.2295455080457032, 'learning_rate/action_model': 7.152210762082623e-06, 'learning_rate/action_query_proj': 7.152210762082623e-06, 'learning_rate/base': 1.7880526905206557e-06, 'epoch': 0.76}) +[2026-06-30 09:12:32,620][starVLA.training.train_starvla][INFO] - Step 1769, Loss: {'train/loss': 0.4483867883682251, 'train/grad_norm_pre_clip': 2.5660669803619385, 'timing/data': 0.00019966601394116879, 'timing/model': 1.2326335571706295, 'learning_rate/action_model': 7.125393279308881e-06, 'learning_rate/action_query_proj': 7.125393279308881e-06, 'learning_rate/base': 1.7813483198272203e-06, 'epoch': 0.76}) +[2026-06-30 09:13:05,308][starVLA.training.train_starvla][INFO] - Step 1770, Loss: {'train/loss': 5.1941111451014876e-05, 'train/grad_norm_pre_clip': 2.217604637145996, 'timing/data': 0.0001961500383913517, 'timing/model': 1.229313839925453, 'learning_rate/action_model': 7.09868652016637e-06, 'learning_rate/action_query_proj': 7.09868652016637e-06, 'learning_rate/base': 1.7746716300415924e-06, 'epoch': 0.76}) +[2026-06-30 09:13:37,934][starVLA.training.train_starvla][INFO] - Step 1771, Loss: {'train/loss': 0.0007641928968951106, 'train/grad_norm_pre_clip': 3.7205569744110107, 'timing/data': 0.0001968909054994583, 'timing/model': 1.231102867051959, 'learning_rate/action_model': 7.072090550551369e-06, 'learning_rate/action_query_proj': 7.072090550551369e-06, 'learning_rate/base': 1.7680226376378422e-06, 'epoch': 0.76}) +[2026-06-30 09:14:10,601][starVLA.training.train_starvla][INFO] - Step 1772, Loss: {'train/loss': 0.00010541974188527092, 'train/grad_norm_pre_clip': 2.1068968772888184, 'timing/data': 0.0002565239556133747, 'timing/model': 1.2323508150875568, 'learning_rate/action_model': 7.045605436086771e-06, 'learning_rate/action_query_proj': 7.045605436086771e-06, 'learning_rate/base': 1.7614013590216927e-06, 'epoch': 0.76}) +[2026-06-30 09:14:43,244][starVLA.training.train_starvla][INFO] - Step 1773, Loss: {'train/loss': 0.28664466738700867, 'train/grad_norm_pre_clip': 3.672922372817993, 'timing/data': 0.00019667111337184906, 'timing/model': 1.2304258460644633, 'learning_rate/action_model': 7.0192312421219805e-06, 'learning_rate/action_query_proj': 7.0192312421219805e-06, 'learning_rate/base': 1.7548078105304951e-06, 'epoch': 0.76}) +[2026-06-30 09:15:15,874][starVLA.training.train_starvla][INFO] - Step 1774, Loss: {'train/loss': 0.0030010908376425505, 'train/grad_norm_pre_clip': 2.7462611198425293, 'timing/data': 0.00026105204597115517, 'timing/model': 1.2321222790051252, 'learning_rate/action_model': 6.992968033732694e-06, 'learning_rate/action_query_proj': 6.992968033732694e-06, 'learning_rate/base': 1.7482420084331735e-06, 'epoch': 0.76}) +[2026-06-30 09:15:48,546][starVLA.training.train_starvla][INFO] - Step 1775, Loss: {'train/loss': 0.44425442814826965, 'train/grad_norm_pre_clip': 2.080237865447998, 'timing/data': 0.00035325600765645504, 'timing/model': 1.2306566769257188, 'learning_rate/action_model': 6.966815875720766e-06, 'learning_rate/action_query_proj': 6.966815875720766e-06, 'learning_rate/base': 1.7417039689301914e-06, 'epoch': 0.76}) +[2026-06-30 09:16:21,197][starVLA.training.train_starvla][INFO] - Step 1776, Loss: {'train/loss': 0.3130955696105957, 'train/grad_norm_pre_clip': 3.3792591094970703, 'timing/data': 0.00019458704628050327, 'timing/model': 1.2322229440324008, 'learning_rate/action_model': 6.9407748326140445e-06, 'learning_rate/action_query_proj': 6.9407748326140445e-06, 'learning_rate/base': 1.7351937081535111e-06, 'epoch': 0.76}) +[2026-06-30 09:16:53,855][starVLA.training.train_starvla][INFO] - Step 1777, Loss: {'train/loss': 0.010561448521912098, 'train/grad_norm_pre_clip': 1.4289970397949219, 'timing/data': 0.00020060897804796696, 'timing/model': 1.2330356989987195, 'learning_rate/action_model': 6.91484496866622e-06, 'learning_rate/action_query_proj': 6.91484496866622e-06, 'learning_rate/base': 1.728711242166555e-06, 'epoch': 0.76}) +[2026-06-30 09:17:26,548][starVLA.training.train_starvla][INFO] - Step 1778, Loss: {'train/loss': 0.10643554478883743, 'train/grad_norm_pre_clip': 2.5639355182647705, 'timing/data': 0.00027766311541199684, 'timing/model': 1.2292931720148772, 'learning_rate/action_model': 6.889026347856656e-06, 'learning_rate/action_query_proj': 6.889026347856656e-06, 'learning_rate/base': 1.722256586964164e-06, 'epoch': 0.77}) +[2026-06-30 09:17:59,170][starVLA.training.train_starvla][INFO] - Step 1779, Loss: {'train/loss': 0.08307883143424988, 'train/grad_norm_pre_clip': 2.5680501461029053, 'timing/data': 0.0002499110996723175, 'timing/model': 1.2317196510266513, 'learning_rate/action_model': 6.863319033890231e-06, 'learning_rate/action_query_proj': 6.863319033890231e-06, 'learning_rate/base': 1.7158297584725577e-06, 'epoch': 0.77}) +[2026-06-30 09:18:31,852][starVLA.training.train_starvla][INFO] - Step 1780, Loss: {'train/loss': 0.00011635197733994573, 'train/grad_norm_pre_clip': 3.234273910522461, 'timing/data': 0.0002191229723393917, 'timing/model': 1.2357489729765803, 'learning_rate/action_model': 6.837723090197178e-06, 'learning_rate/action_query_proj': 6.837723090197178e-06, 'learning_rate/base': 1.7094307725492944e-06, 'epoch': 0.77}) +[2026-06-30 09:19:04,568][starVLA.training.train_starvla][INFO] - Step 1781, Loss: {'train/loss': 0.0021477064583450556, 'train/grad_norm_pre_clip': 3.914527177810669, 'timing/data': 0.0002426279243081808, 'timing/model': 1.2343612480908632, 'learning_rate/action_model': 6.812238579932956e-06, 'learning_rate/action_query_proj': 6.812238579932956e-06, 'learning_rate/base': 1.703059644983239e-06, 'epoch': 0.77}) +[2026-06-30 09:19:37,238][starVLA.training.train_starvla][INFO] - Step 1782, Loss: {'train/loss': 0.1841658055782318, 'train/grad_norm_pre_clip': 2.076533555984497, 'timing/data': 0.00028678099624812603, 'timing/model': 1.2304119290784001, 'learning_rate/action_model': 6.786865565978054e-06, 'learning_rate/action_query_proj': 6.786865565978054e-06, 'learning_rate/base': 1.6967163914945136e-06, 'epoch': 0.77}) +[2026-06-30 09:20:09,929][starVLA.training.train_starvla][INFO] - Step 1783, Loss: {'train/loss': 0.013433527201414108, 'train/grad_norm_pre_clip': 2.025444269180298, 'timing/data': 0.0004858560860157013, 'timing/model': 1.2299421238712966, 'learning_rate/action_model': 6.761604110937861e-06, 'learning_rate/action_query_proj': 6.761604110937861e-06, 'learning_rate/base': 1.6904010277344652e-06, 'epoch': 0.77}) +[2026-06-30 09:20:42,611][starVLA.training.train_starvla][INFO] - Step 1784, Loss: {'train/loss': 2.6226007321383804e-06, 'train/grad_norm_pre_clip': 2.3867428302764893, 'timing/data': 0.000407207990065217, 'timing/model': 1.230571203166619, 'learning_rate/action_model': 6.736454277142502e-06, 'learning_rate/action_query_proj': 6.736454277142502e-06, 'learning_rate/base': 1.6841135692856255e-06, 'epoch': 0.77}) +[2026-06-30 09:21:15,304][starVLA.training.train_starvla][INFO] - Step 1785, Loss: {'train/loss': 0.022414591163396835, 'train/grad_norm_pre_clip': 2.7734036445617676, 'timing/data': 0.00018966803327202797, 'timing/model': 1.2331226968672127, 'learning_rate/action_model': 6.711416126646712e-06, 'learning_rate/action_query_proj': 6.711416126646712e-06, 'learning_rate/base': 1.677854031661678e-06, 'epoch': 0.77}) +[2026-06-30 09:21:47,977][starVLA.training.train_starvla][INFO] - Step 1786, Loss: {'train/loss': 5.841226538905175e-06, 'train/grad_norm_pre_clip': 2.9588818550109863, 'timing/data': 0.00039864215068519115, 'timing/model': 1.2296822890639305, 'learning_rate/action_model': 6.686489721229616e-06, 'learning_rate/action_query_proj': 6.686489721229616e-06, 'learning_rate/base': 1.671622430307404e-06, 'epoch': 0.77}) +[2026-06-30 09:22:20,672][starVLA.training.train_starvla][INFO] - Step 1787, Loss: {'train/loss': 0.0018900232389569283, 'train/grad_norm_pre_clip': 3.1964263916015625, 'timing/data': 0.0002266170922666788, 'timing/model': 1.232992673991248, 'learning_rate/action_model': 6.661675122394651e-06, 'learning_rate/action_query_proj': 6.661675122394651e-06, 'learning_rate/base': 1.6654187805986627e-06, 'epoch': 0.77}) +[2026-06-30 09:22:53,318][starVLA.training.train_starvla][INFO] - Step 1788, Loss: {'train/loss': 9.14924567041453e-06, 'train/grad_norm_pre_clip': 1.7369719743728638, 'timing/data': 0.0003287300933152437, 'timing/model': 1.2282815091311932, 'learning_rate/action_model': 6.636972391369374e-06, 'learning_rate/action_query_proj': 6.636972391369374e-06, 'learning_rate/base': 1.6592430978423434e-06, 'epoch': 0.77}) +[2026-06-30 09:23:25,984][starVLA.training.train_starvla][INFO] - Step 1789, Loss: {'train/loss': 0.44388341903686523, 'train/grad_norm_pre_clip': 1.7596933841705322, 'timing/data': 0.0002558829728513956, 'timing/model': 1.2348923389799893, 'learning_rate/action_model': 6.612381589105311e-06, 'learning_rate/action_query_proj': 6.612381589105311e-06, 'learning_rate/base': 1.6530953972763278e-06, 'epoch': 0.77}) +[2026-06-30 09:23:58,645][starVLA.training.train_starvla][INFO] - Step 1790, Loss: {'train/loss': 9.582775965100154e-05, 'train/grad_norm_pre_clip': 3.4590542316436768, 'timing/data': 0.00020243204198777676, 'timing/model': 1.2339296350255609, 'learning_rate/action_model': 6.587902776277825e-06, 'learning_rate/action_query_proj': 6.587902776277825e-06, 'learning_rate/base': 1.6469756940694563e-06, 'epoch': 0.77}) +[2026-06-30 09:24:31,264][starVLA.training.train_starvla][INFO] - Step 1791, Loss: {'train/loss': 0.00012264940596651286, 'train/grad_norm_pre_clip': 2.354083299636841, 'timing/data': 0.0002537581603974104, 'timing/model': 1.2328183590434492, 'learning_rate/action_model': 6.5635360132859575e-06, 'learning_rate/action_query_proj': 6.5635360132859575e-06, 'learning_rate/base': 1.6408840033214894e-06, 'epoch': 0.77}) +[2026-06-30 09:25:03,909][starVLA.training.train_starvla][INFO] - Step 1792, Loss: {'train/loss': 0.007964633405208588, 'train/grad_norm_pre_clip': 2.2224016189575195, 'timing/data': 0.00030219019390642643, 'timing/model': 1.2263340221252292, 'learning_rate/action_model': 6.5392813602522764e-06, 'learning_rate/action_query_proj': 6.5392813602522764e-06, 'learning_rate/base': 1.6348203400630691e-06, 'epoch': 0.77}) +[2026-06-30 09:25:36,562][starVLA.training.train_starvla][INFO] - Step 1793, Loss: {'train/loss': 0.3921782076358795, 'train/grad_norm_pre_clip': 1.8319201469421387, 'timing/data': 0.00020447513088583946, 'timing/model': 1.2316262829117477, 'learning_rate/action_model': 6.5151388770227225e-06, 'learning_rate/action_query_proj': 6.5151388770227225e-06, 'learning_rate/base': 1.6287847192556806e-06, 'epoch': 0.77}) +[2026-06-30 09:26:09,213][starVLA.training.train_starvla][INFO] - Step 1794, Loss: {'train/loss': 0.27306410670280457, 'train/grad_norm_pre_clip': 2.825449228286743, 'timing/data': 0.00019592000171542168, 'timing/model': 1.2297206490766257, 'learning_rate/action_model': 6.491108623166463e-06, 'learning_rate/action_query_proj': 6.491108623166463e-06, 'learning_rate/base': 1.6227771557916158e-06, 'epoch': 0.77}) +[2026-06-30 09:26:41,872][starVLA.training.train_starvla][INFO] - Step 1795, Loss: {'train/loss': 0.2914080321788788, 'train/grad_norm_pre_clip': 2.5337769985198975, 'timing/data': 0.00019614002667367458, 'timing/model': 1.232490224065259, 'learning_rate/action_model': 6.4671906579757786e-06, 'learning_rate/action_query_proj': 6.4671906579757786e-06, 'learning_rate/base': 1.6167976644939446e-06, 'epoch': 0.77}) +[2026-06-30 09:27:14,569][starVLA.training.train_starvla][INFO] - Step 1796, Loss: {'train/loss': 0.007322109304368496, 'train/grad_norm_pre_clip': 2.2396023273468018, 'timing/data': 0.00019358494319021702, 'timing/model': 1.2318942011334002, 'learning_rate/action_model': 6.443385040465865e-06, 'learning_rate/action_query_proj': 6.443385040465865e-06, 'learning_rate/base': 1.6108462601164662e-06, 'epoch': 0.77}) +[2026-06-30 09:27:47,218][starVLA.training.train_starvla][INFO] - Step 1797, Loss: {'train/loss': 0.0018520914018154144, 'train/grad_norm_pre_clip': 1.2465611696243286, 'timing/data': 0.00019052904099225998, 'timing/model': 1.2277485311497003, 'learning_rate/action_model': 6.419691829374709e-06, 'learning_rate/action_query_proj': 6.419691829374709e-06, 'learning_rate/base': 1.6049229573436773e-06, 'epoch': 0.77}) +[2026-06-30 09:28:19,872][starVLA.training.train_starvla][INFO] - Step 1798, Loss: {'train/loss': 0.014717922545969486, 'train/grad_norm_pre_clip': 3.749692916870117, 'timing/data': 0.00018958793953061104, 'timing/model': 1.2329759011045098, 'learning_rate/action_model': 6.396111083162953e-06, 'learning_rate/action_query_proj': 6.396111083162953e-06, 'learning_rate/base': 1.5990277707907383e-06, 'epoch': 0.77}) +[2026-06-30 09:28:52,499][starVLA.training.train_starvla][INFO] - Step 1799, Loss: {'train/loss': 3.766796726267785e-05, 'train/grad_norm_pre_clip': 3.243655204772949, 'timing/data': 0.00020498712547123432, 'timing/model': 1.2216891830321401, 'learning_rate/action_model': 6.37264286001376e-06, 'learning_rate/action_query_proj': 6.37264286001376e-06, 'learning_rate/base': 1.59316071500344e-06, 'epoch': 0.77}) +[2026-06-30 09:29:25,185][starVLA.training.train_starvla][INFO] - Step 1800, Loss: {'train/loss': 0.044083308428525925, 'train/grad_norm_pre_clip': 1.097751259803772, 'timing/data': 0.00040300097316503525, 'timing/model': 1.2376248720102012, 'learning_rate/action_model': 6.349287217832631e-06, 'learning_rate/action_query_proj': 6.349287217832631e-06, 'learning_rate/base': 1.5873218044581577e-06, 'epoch': 0.77}) +[2026-06-30 09:29:57,836][starVLA.training.train_starvla][INFO] - Step 1801, Loss: {'train/loss': 0.013237380422651768, 'train/grad_norm_pre_clip': 1.8071619272232056, 'timing/data': 0.00019644014537334442, 'timing/model': 1.2323054049629718, 'learning_rate/action_model': 6.326044214247284e-06, 'learning_rate/action_query_proj': 6.326044214247284e-06, 'learning_rate/base': 1.581511053561821e-06, 'epoch': 0.77}) +[2026-06-30 09:30:30,503][starVLA.training.train_starvla][INFO] - Step 1802, Loss: {'train/loss': 0.009548068977892399, 'train/grad_norm_pre_clip': 2.4589641094207764, 'timing/data': 0.00019156117923557758, 'timing/model': 1.2306578939314932, 'learning_rate/action_model': 6.302913906607521e-06, 'learning_rate/action_query_proj': 6.302913906607521e-06, 'learning_rate/base': 1.5757284766518803e-06, 'epoch': 0.78}) +[2026-06-30 09:31:03,164][starVLA.training.train_starvla][INFO] - Step 1803, Loss: {'train/loss': 0.004187143407762051, 'train/grad_norm_pre_clip': 3.3338723182678223, 'timing/data': 0.0003001859877258539, 'timing/model': 1.2276949831284583, 'learning_rate/action_model': 6.279896351985093e-06, 'learning_rate/action_query_proj': 6.279896351985093e-06, 'learning_rate/base': 1.5699740879962733e-06, 'epoch': 0.78}) +[2026-06-30 09:31:35,827][starVLA.training.train_starvla][INFO] - Step 1804, Loss: {'train/loss': 0.05645319074392319, 'train/grad_norm_pre_clip': 1.3718743324279785, 'timing/data': 0.00019818381406366825, 'timing/model': 1.2273035009857267, 'learning_rate/action_model': 6.2569916071735e-06, 'learning_rate/action_query_proj': 6.2569916071735e-06, 'learning_rate/base': 1.564247901793375e-06, 'epoch': 0.78}) +[2026-06-30 09:32:08,475][starVLA.training.train_starvla][INFO] - Step 1805, Loss: {'train/loss': 0.055669210851192474, 'train/grad_norm_pre_clip': 2.756779193878174, 'timing/data': 0.00034968997351825237, 'timing/model': 1.2307663669344038, 'learning_rate/action_model': 6.234199728687943e-06, 'learning_rate/action_query_proj': 6.234199728687943e-06, 'learning_rate/base': 1.5585499321719857e-06, 'epoch': 0.78}) +[2026-06-30 09:32:41,169][starVLA.training.train_starvla][INFO] - Step 1806, Loss: {'train/loss': 0.21308931708335876, 'train/grad_norm_pre_clip': 2.7994110584259033, 'timing/data': 0.00020215101540088654, 'timing/model': 1.2294026489835232, 'learning_rate/action_model': 6.211520772765114e-06, 'learning_rate/action_query_proj': 6.211520772765114e-06, 'learning_rate/base': 1.5528801931912785e-06, 'epoch': 0.78}) +[2026-06-30 09:33:13,818][starVLA.training.train_starvla][INFO] - Step 1807, Loss: {'train/loss': 3.02777043543756e-05, 'train/grad_norm_pre_clip': 1.7154642343521118, 'timing/data': 0.00020939507521688938, 'timing/model': 1.2226295571308583, 'learning_rate/action_model': 6.188954795363075e-06, 'learning_rate/action_query_proj': 6.188954795363075e-06, 'learning_rate/base': 1.5472386988407687e-06, 'epoch': 0.78}) +[2026-06-30 09:33:46,430][starVLA.training.train_starvla][INFO] - Step 1808, Loss: {'train/loss': 0.5419678688049316, 'train/grad_norm_pre_clip': 2.7932441234588623, 'timing/data': 0.00020532705821096897, 'timing/model': 1.228584931930527, 'learning_rate/action_model': 6.166501852161138e-06, 'learning_rate/action_query_proj': 6.166501852161138e-06, 'learning_rate/base': 1.5416254630402846e-06, 'epoch': 0.78}) +[2026-06-30 09:34:19,078][starVLA.training.train_starvla][INFO] - Step 1809, Loss: {'train/loss': 0.005908128339797258, 'train/grad_norm_pre_clip': 1.6482908725738525, 'timing/data': 0.00019227201119065285, 'timing/model': 1.2326657299418002, 'learning_rate/action_model': 6.144161998559711e-06, 'learning_rate/action_query_proj': 6.144161998559711e-06, 'learning_rate/base': 1.5360404996399277e-06, 'epoch': 0.78}) +[2026-06-30 09:34:51,739][starVLA.training.train_starvla][INFO] - Step 1810, Loss: {'train/loss': 0.403160959482193, 'train/grad_norm_pre_clip': 1.69125497341156, 'timing/data': 0.00027717300690710545, 'timing/model': 1.2310060909949243, 'learning_rate/action_model': 6.121935289680153e-06, 'learning_rate/action_query_proj': 6.121935289680153e-06, 'learning_rate/base': 1.5304838224200383e-06, 'epoch': 0.78}) +[2026-06-30 09:35:24,408][starVLA.training.train_starvla][INFO] - Step 1811, Loss: {'train/loss': 2.4437867978122085e-06, 'train/grad_norm_pre_clip': 2.3985979557037354, 'timing/data': 0.00029576802626252174, 'timing/model': 1.2307463721372187, 'learning_rate/action_model': 6.099821780364671e-06, 'learning_rate/action_query_proj': 6.099821780364671e-06, 'learning_rate/base': 1.5249554450911677e-06, 'epoch': 0.78}) +[2026-06-30 09:35:56,990][starVLA.training.train_starvla][INFO] - Step 1812, Loss: {'train/loss': 2.741699427133426e-05, 'train/grad_norm_pre_clip': 3.9285032749176025, 'timing/data': 0.00034358794800937176, 'timing/model': 1.2288304879330099, 'learning_rate/action_model': 6.077821525176133e-06, 'learning_rate/action_query_proj': 6.077821525176133e-06, 'learning_rate/base': 1.5194553812940334e-06, 'epoch': 0.78}) +[2026-06-30 09:36:29,647][starVLA.training.train_starvla][INFO] - Step 1813, Loss: {'train/loss': 0.0673196017742157, 'train/grad_norm_pre_clip': 2.8943769931793213, 'timing/data': 0.000206278869882226, 'timing/model': 1.2315689318347722, 'learning_rate/action_model': 6.055934578398005e-06, 'learning_rate/action_query_proj': 6.055934578398005e-06, 'learning_rate/base': 1.5139836445995013e-06, 'epoch': 0.78}) +[2026-06-30 09:37:02,316][starVLA.training.train_starvla][INFO] - Step 1814, Loss: {'train/loss': 0.027486592531204224, 'train/grad_norm_pre_clip': 3.3064627647399902, 'timing/data': 0.00022414186969399452, 'timing/model': 1.2300017639063299, 'learning_rate/action_model': 6.034160994034132e-06, 'learning_rate/action_query_proj': 6.034160994034132e-06, 'learning_rate/base': 1.508540248508533e-06, 'epoch': 0.78}) +[2026-06-30 09:37:34,956][starVLA.training.train_starvla][INFO] - Step 1815, Loss: {'train/loss': 0.12745678424835205, 'train/grad_norm_pre_clip': 2.2378599643707275, 'timing/data': 0.0002067300956696272, 'timing/model': 1.2306532710790634, 'learning_rate/action_model': 6.012500825808682e-06, 'learning_rate/action_query_proj': 6.012500825808682e-06, 'learning_rate/base': 1.5031252064521706e-06, 'epoch': 0.78}) +[2026-06-30 09:38:07,824][starVLA.training.train_starvla][INFO] - Step 1816, Loss: {'train/loss': 0.00023425887047778815, 'train/grad_norm_pre_clip': 1.8308206796646118, 'timing/data': 0.0002847970463335514, 'timing/model': 1.4379032501019537, 'learning_rate/action_model': 5.990954127165954e-06, 'learning_rate/action_query_proj': 5.990954127165954e-06, 'learning_rate/base': 1.4977385317914885e-06, 'epoch': 0.78}) +[2026-06-30 09:38:40,511][starVLA.training.train_starvla][INFO] - Step 1817, Loss: {'train/loss': 1.22486144391587e-05, 'train/grad_norm_pre_clip': 3.0371696949005127, 'timing/data': 0.00023561413399875164, 'timing/model': 1.2282076328992844, 'learning_rate/action_model': 5.9695209512703006e-06, 'learning_rate/action_query_proj': 5.9695209512703006e-06, 'learning_rate/base': 1.4923802378175751e-06, 'epoch': 0.78}) +[2026-06-30 09:39:13,183][starVLA.training.train_starvla][INFO] - Step 1818, Loss: {'train/loss': 0.014780594035983086, 'train/grad_norm_pre_clip': 1.4660828113555908, 'timing/data': 0.00033018295653164387, 'timing/model': 1.2312120059505105, 'learning_rate/action_model': 5.948201351005947e-06, 'learning_rate/action_query_proj': 5.948201351005947e-06, 'learning_rate/base': 1.4870503377514866e-06, 'epoch': 0.78}) +[2026-06-30 09:39:45,805][starVLA.training.train_starvla][INFO] - Step 1819, Loss: {'train/loss': 0.2980675995349884, 'train/grad_norm_pre_clip': 4.277560234069824, 'timing/data': 0.0002014501951634884, 'timing/model': 1.2270997560117394, 'learning_rate/action_model': 5.926995378976878e-06, 'learning_rate/action_query_proj': 5.926995378976878e-06, 'learning_rate/base': 1.4817488447442195e-06, 'epoch': 0.78}) +[2026-06-30 09:40:18,448][starVLA.training.train_starvla][INFO] - Step 1820, Loss: {'train/loss': 0.03555670008063316, 'train/grad_norm_pre_clip': 2.074075222015381, 'timing/data': 0.00020439503714442253, 'timing/model': 1.2322362600825727, 'learning_rate/action_model': 5.905903087506733e-06, 'learning_rate/action_query_proj': 5.905903087506733e-06, 'learning_rate/base': 1.4764757718766832e-06, 'epoch': 0.78}) +[2026-06-30 09:40:51,061][starVLA.training.train_starvla][INFO] - Step 1821, Loss: {'train/loss': 0.8559864163398743, 'train/grad_norm_pre_clip': 3.9666900634765625, 'timing/data': 0.0002169688232243061, 'timing/model': 1.2326831279788166, 'learning_rate/action_model': 5.884924528638644e-06, 'learning_rate/action_query_proj': 5.884924528638644e-06, 'learning_rate/base': 1.471231132159661e-06, 'epoch': 0.78}) +[2026-06-30 09:41:23,671][starVLA.training.train_starvla][INFO] - Step 1822, Loss: {'train/loss': 0.004461872391402721, 'train/grad_norm_pre_clip': 4.955499172210693, 'timing/data': 0.00034671416506171227, 'timing/model': 1.2334196621086448, 'learning_rate/action_model': 5.864059754135108e-06, 'learning_rate/action_query_proj': 5.864059754135108e-06, 'learning_rate/base': 1.466014938533777e-06, 'epoch': 0.78}) +[2026-06-30 09:41:56,288][starVLA.training.train_starvla][INFO] - Step 1823, Loss: {'train/loss': 0.004565000534057617, 'train/grad_norm_pre_clip': 2.093365430831909, 'timing/data': 0.0002649701200425625, 'timing/model': 1.2274104729294777, 'learning_rate/action_model': 5.843308815477897e-06, 'learning_rate/action_query_proj': 5.843308815477897e-06, 'learning_rate/base': 1.4608272038694742e-06, 'epoch': 0.78}) +[2026-06-30 09:42:28,957][starVLA.training.train_starvla][INFO] - Step 1824, Loss: {'train/loss': 0.00011721672490239143, 'train/grad_norm_pre_clip': 3.398369073867798, 'timing/data': 0.0002593989484012127, 'timing/model': 1.2259211488999426, 'learning_rate/action_model': 5.8226717638678745e-06, 'learning_rate/action_query_proj': 5.8226717638678745e-06, 'learning_rate/base': 1.4556679409669686e-06, 'epoch': 0.78}) +[2026-06-30 09:43:01,596][starVLA.training.train_starvla][INFO] - Step 1825, Loss: {'train/loss': 6.258463599806419e-06, 'train/grad_norm_pre_clip': 2.413468837738037, 'timing/data': 0.0003068288788199425, 'timing/model': 1.2317483939696103, 'learning_rate/action_model': 5.8021486502249355e-06, 'learning_rate/action_query_proj': 5.8021486502249355e-06, 'learning_rate/base': 1.4505371625562339e-06, 'epoch': 0.79}) +[2026-06-30 09:43:34,271][starVLA.training.train_starvla][INFO] - Step 1826, Loss: {'train/loss': 0.013547616079449654, 'train/grad_norm_pre_clip': 2.3092098236083984, 'timing/data': 0.00026159314438700676, 'timing/model': 1.2278566369786859, 'learning_rate/action_model': 5.781739525187793e-06, 'learning_rate/action_query_proj': 5.781739525187793e-06, 'learning_rate/base': 1.4454348812969483e-06, 'epoch': 0.79}) +[2026-06-30 09:44:06,898][starVLA.training.train_starvla][INFO] - Step 1827, Loss: {'train/loss': 0.020470408722758293, 'train/grad_norm_pre_clip': 1.8978434801101685, 'timing/data': 0.00022910209372639656, 'timing/model': 1.2282227110117674, 'learning_rate/action_model': 5.761444439113964e-06, 'learning_rate/action_query_proj': 5.761444439113964e-06, 'learning_rate/base': 1.440361109778491e-06, 'epoch': 0.79}) +[2026-06-30 09:44:39,556][starVLA.training.train_starvla][INFO] - Step 1828, Loss: {'train/loss': 0.19845803081989288, 'train/grad_norm_pre_clip': 1.6040152311325073, 'timing/data': 0.00020598899573087692, 'timing/model': 1.2298761229030788, 'learning_rate/action_model': 5.7412634420795336e-06, 'learning_rate/action_query_proj': 5.7412634420795336e-06, 'learning_rate/base': 1.4353158605198834e-06, 'epoch': 0.79}) +[2026-06-30 09:45:12,218][starVLA.training.train_starvla][INFO] - Step 1829, Loss: {'train/loss': 0.00028510979609563947, 'train/grad_norm_pre_clip': 2.097058057785034, 'timing/data': 0.0004484450910240412, 'timing/model': 1.2326048330869526, 'learning_rate/action_model': 5.72119658387912e-06, 'learning_rate/action_query_proj': 5.72119658387912e-06, 'learning_rate/base': 1.43029914596978e-06, 'epoch': 0.79}) +[2026-06-30 09:45:44,884][starVLA.training.train_starvla][INFO] - Step 1830, Loss: {'train/loss': 0.0386384092271328, 'train/grad_norm_pre_clip': 2.8262670040130615, 'timing/data': 0.0002340411301702261, 'timing/model': 1.2323953420855105, 'learning_rate/action_model': 5.701243914025692e-06, 'learning_rate/action_query_proj': 5.701243914025692e-06, 'learning_rate/base': 1.425310978506423e-06, 'epoch': 0.79}) +[2026-06-30 09:46:17,516][starVLA.training.train_starvla][INFO] - Step 1831, Loss: {'train/loss': 0.06492314487695694, 'train/grad_norm_pre_clip': 2.0399670600891113, 'timing/data': 0.0003180198837071657, 'timing/model': 1.2324507387820631, 'learning_rate/action_model': 5.6814054817504824e-06, 'learning_rate/action_query_proj': 5.6814054817504824e-06, 'learning_rate/base': 1.4203513704376206e-06, 'epoch': 0.79}) +[2026-06-30 09:46:50,136][starVLA.training.train_starvla][INFO] - Step 1832, Loss: {'train/loss': 3.2782472771941684e-06, 'train/grad_norm_pre_clip': 1.958715558052063, 'timing/data': 0.0002561530563980341, 'timing/model': 1.230345570947975, 'learning_rate/action_model': 5.661681336002853e-06, 'learning_rate/action_query_proj': 5.661681336002853e-06, 'learning_rate/base': 1.4154203340007133e-06, 'epoch': 0.79}) +[2026-06-30 09:47:22,780][starVLA.training.train_starvla][INFO] - Step 1833, Loss: {'train/loss': 0.001351216109469533, 'train/grad_norm_pre_clip': 1.9978333711624146, 'timing/data': 0.00021380302496254444, 'timing/model': 1.2290651581715792, 'learning_rate/action_model': 5.642071525450173e-06, 'learning_rate/action_query_proj': 5.642071525450173e-06, 'learning_rate/base': 1.4105178813625433e-06, 'epoch': 0.79}) +[2026-06-30 09:47:55,412][starVLA.training.train_starvla][INFO] - Step 1834, Loss: {'train/loss': 0.07307512313127518, 'train/grad_norm_pre_clip': 2.8067238330841064, 'timing/data': 0.00033930991776287556, 'timing/model': 1.2289978328626603, 'learning_rate/action_model': 5.62257609847769e-06, 'learning_rate/action_query_proj': 5.62257609847769e-06, 'learning_rate/base': 1.4056440246194225e-06, 'epoch': 0.79}) +[2026-06-30 09:48:28,045][starVLA.training.train_starvla][INFO] - Step 1835, Loss: {'train/loss': 3.278249096183572e-06, 'train/grad_norm_pre_clip': 1.487243890762329, 'timing/data': 0.00023189792409539223, 'timing/model': 1.2281697939615697, 'learning_rate/action_model': 5.603195103188451e-06, 'learning_rate/action_query_proj': 5.603195103188451e-06, 'learning_rate/base': 1.4007987757971127e-06, 'epoch': 0.79}) +[2026-06-30 09:49:00,682][starVLA.training.train_starvla][INFO] - Step 1836, Loss: {'train/loss': 0.8108603954315186, 'train/grad_norm_pre_clip': 2.2811431884765625, 'timing/data': 0.0002675249706953764, 'timing/model': 1.2305901208892465, 'learning_rate/action_model': 5.5839285874031196e-06, 'learning_rate/action_query_proj': 5.5839285874031196e-06, 'learning_rate/base': 1.3959821468507799e-06, 'epoch': 0.79}) +[2026-06-30 09:49:33,312][starVLA.training.train_starvla][INFO] - Step 1837, Loss: {'train/loss': 0.08321505039930344, 'train/grad_norm_pre_clip': 3.388542413711548, 'timing/data': 0.0003088931553065777, 'timing/model': 1.2272294438444078, 'learning_rate/action_model': 5.564776598659911e-06, 'learning_rate/action_query_proj': 5.564776598659911e-06, 'learning_rate/base': 1.3911941496649778e-06, 'epoch': 0.79}) +[2026-06-30 09:50:05,971][starVLA.training.train_starvla][INFO] - Step 1838, Loss: {'train/loss': 0.00010026575182564557, 'train/grad_norm_pre_clip': 3.4592180252075195, 'timing/data': 0.0003072188701480627, 'timing/model': 1.2284807260148227, 'learning_rate/action_model': 5.545739184214458e-06, 'learning_rate/action_query_proj': 5.545739184214458e-06, 'learning_rate/base': 1.3864347960536144e-06, 'epoch': 0.79}) +[2026-06-30 09:50:38,633][starVLA.training.train_starvla][INFO] - Step 1839, Loss: {'train/loss': 3.0277808036771603e-05, 'train/grad_norm_pre_clip': 1.0948290824890137, 'timing/data': 0.00028048898093402386, 'timing/model': 1.228837390895933, 'learning_rate/action_model': 5.526816391039688e-06, 'learning_rate/action_query_proj': 5.526816391039688e-06, 'learning_rate/base': 1.381704097759922e-06, 'epoch': 0.79}) +[2026-06-30 09:51:11,335][starVLA.training.train_starvla][INFO] - Step 1840, Loss: {'train/loss': 0.13036055862903595, 'train/grad_norm_pre_clip': 2.871227502822876, 'timing/data': 0.0002609721850603819, 'timing/model': 1.2284784158691764, 'learning_rate/action_model': 5.508008265825708e-06, 'learning_rate/action_query_proj': 5.508008265825708e-06, 'learning_rate/base': 1.377002066456427e-06, 'epoch': 0.79}) +[2026-06-30 09:51:43,908][starVLA.training.train_starvla][INFO] - Step 1841, Loss: {'train/loss': 0.03782166913151741, 'train/grad_norm_pre_clip': 3.0185036659240723, 'timing/data': 0.00027275504544377327, 'timing/model': 1.2330088920425624, 'learning_rate/action_model': 5.489314854979701e-06, 'learning_rate/action_query_proj': 5.489314854979701e-06, 'learning_rate/base': 1.3723287137449253e-06, 'epoch': 0.79}) +[2026-06-30 09:52:16,595][starVLA.training.train_starvla][INFO] - Step 1842, Loss: {'train/loss': 0.0012163728242740035, 'train/grad_norm_pre_clip': 2.2894773483276367, 'timing/data': 0.0002590389922261238, 'timing/model': 1.2302957060746849, 'learning_rate/action_model': 5.47073620462579e-06, 'learning_rate/action_query_proj': 5.47073620462579e-06, 'learning_rate/base': 1.3676840511564475e-06, 'epoch': 0.79}) +[2026-06-30 09:52:49,190][starVLA.training.train_starvla][INFO] - Step 1843, Loss: {'train/loss': 0.08125074207782745, 'train/grad_norm_pre_clip': 1.613913893699646, 'timing/data': 0.00019793305546045303, 'timing/model': 1.2293320200406015, 'learning_rate/action_model': 5.45227236060497e-06, 'learning_rate/action_query_proj': 5.45227236060497e-06, 'learning_rate/base': 1.3630680901512426e-06, 'epoch': 0.79}) +[2026-06-30 09:53:21,838][starVLA.training.train_starvla][INFO] - Step 1844, Loss: {'train/loss': 0.463084876537323, 'train/grad_norm_pre_clip': 3.56007981300354, 'timing/data': 0.00020756199955940247, 'timing/model': 1.2326363830361515, 'learning_rate/action_model': 5.433923368474921e-06, 'learning_rate/action_query_proj': 5.433923368474921e-06, 'learning_rate/base': 1.3584808421187303e-06, 'epoch': 0.79}) +[2026-06-30 09:53:54,521][starVLA.training.train_starvla][INFO] - Step 1845, Loss: {'train/loss': 0.2833336293697357, 'train/grad_norm_pre_clip': 1.9853233098983765, 'timing/data': 0.0001917220652103424, 'timing/model': 1.231790428981185, 'learning_rate/action_model': 5.415689273509961e-06, 'learning_rate/action_query_proj': 5.415689273509961e-06, 'learning_rate/base': 1.3539223183774903e-06, 'epoch': 0.79}) +[2026-06-30 09:54:27,186][starVLA.training.train_starvla][INFO] - Step 1846, Loss: {'train/loss': 0.720805287361145, 'train/grad_norm_pre_clip': 2.9298739433288574, 'timing/data': 0.0002681158948689699, 'timing/model': 1.2302127238363028, 'learning_rate/action_model': 5.397570120700912e-06, 'learning_rate/action_query_proj': 5.397570120700912e-06, 'learning_rate/base': 1.349392530175228e-06, 'epoch': 0.79}) +[2026-06-30 09:54:59,858][starVLA.training.train_starvla][INFO] - Step 1847, Loss: {'train/loss': 3.6358737816044595e-06, 'train/grad_norm_pre_clip': 3.1883111000061035, 'timing/data': 0.0002052679192274809, 'timing/model': 1.2347186738625169, 'learning_rate/action_model': 5.379565954754982e-06, 'learning_rate/action_query_proj': 5.379565954754982e-06, 'learning_rate/base': 1.3448914886887454e-06, 'epoch': 0.79}) +[2026-06-30 09:55:32,528][starVLA.training.train_starvla][INFO] - Step 1848, Loss: {'train/loss': 0.1372404247522354, 'train/grad_norm_pre_clip': 1.8559592962265015, 'timing/data': 0.00024849805049598217, 'timing/model': 1.231900820042938, 'learning_rate/action_model': 5.361676820095649e-06, 'learning_rate/action_query_proj': 5.361676820095649e-06, 'learning_rate/base': 1.3404192050239122e-06, 'epoch': 0.8}) +[2026-06-30 09:56:05,207][starVLA.training.train_starvla][INFO] - Step 1849, Loss: {'train/loss': 0.00836509931832552, 'train/grad_norm_pre_clip': 3.069678544998169, 'timing/data': 0.00023288908414542675, 'timing/model': 1.2279071281664073, 'learning_rate/action_model': 5.343902760862592e-06, 'learning_rate/action_query_proj': 5.343902760862592e-06, 'learning_rate/base': 1.335975690215648e-06, 'epoch': 0.8}) +[2026-06-30 09:56:37,832][starVLA.training.train_starvla][INFO] - Step 1850, Loss: {'train/loss': 0.43098846077919006, 'train/grad_norm_pre_clip': 2.2425878047943115, 'timing/data': 0.00028561800718307495, 'timing/model': 1.2303101308643818, 'learning_rate/action_model': 5.326243820911525e-06, 'learning_rate/action_query_proj': 5.326243820911525e-06, 'learning_rate/base': 1.3315609552278813e-06, 'epoch': 0.8}) +[2026-06-30 09:57:10,543][starVLA.training.train_starvla][INFO] - Step 1851, Loss: {'train/loss': 0.0014258571900427341, 'train/grad_norm_pre_clip': 1.5346152782440186, 'timing/data': 0.00020651891827583313, 'timing/model': 1.2310479239095002, 'learning_rate/action_model': 5.308700043814127e-06, 'learning_rate/action_query_proj': 5.308700043814127e-06, 'learning_rate/base': 1.3271750109535318e-06, 'epoch': 0.8}) +[2026-06-30 09:57:43,167][starVLA.training.train_starvla][INFO] - Step 1852, Loss: {'train/loss': 0.0002336041652597487, 'train/grad_norm_pre_clip': 2.822234630584717, 'timing/data': 0.00020006694830954075, 'timing/model': 1.2263745900709182, 'learning_rate/action_model': 5.291271472857921e-06, 'learning_rate/action_query_proj': 5.291271472857921e-06, 'learning_rate/base': 1.3228178682144803e-06, 'epoch': 0.8}) +[2026-06-30 09:58:15,823][starVLA.training.train_starvla][INFO] - Step 1853, Loss: {'train/loss': 0.0003409931669011712, 'train/grad_norm_pre_clip': 1.7324305772781372, 'timing/data': 0.0002631561364978552, 'timing/model': 1.2314182131085545, 'learning_rate/action_model': 5.273958151046189e-06, 'learning_rate/action_query_proj': 5.273958151046189e-06, 'learning_rate/base': 1.3184895377615472e-06, 'epoch': 0.8}) +[2026-06-30 09:58:48,481][starVLA.training.train_starvla][INFO] - Step 1854, Loss: {'train/loss': 5.1314811571501195e-05, 'train/grad_norm_pre_clip': 2.8702051639556885, 'timing/data': 0.00026151305064558983, 'timing/model': 1.2295171779114753, 'learning_rate/action_model': 5.2567601210978135e-06, 'learning_rate/action_query_proj': 5.2567601210978135e-06, 'learning_rate/base': 1.3141900302744534e-06, 'epoch': 0.8}) +[2026-06-30 09:59:21,136][starVLA.training.train_starvla][INFO] - Step 1855, Loss: {'train/loss': 0.0007518358179368079, 'train/grad_norm_pre_clip': 2.7699007987976074, 'timing/data': 0.00023250887170433998, 'timing/model': 1.2280900278128684, 'learning_rate/action_model': 5.2396774254472314e-06, 'learning_rate/action_query_proj': 5.2396774254472314e-06, 'learning_rate/base': 1.3099193563618079e-06, 'epoch': 0.8}) +[2026-06-30 09:59:53,815][starVLA.training.train_starvla][INFO] - Step 1856, Loss: {'train/loss': 0.09372448921203613, 'train/grad_norm_pre_clip': 3.289485454559326, 'timing/data': 0.00026808492839336395, 'timing/model': 1.2301590209826827, 'learning_rate/action_model': 5.222710106244293e-06, 'learning_rate/action_query_proj': 5.222710106244293e-06, 'learning_rate/base': 1.3056775265610732e-06, 'epoch': 0.8}) +[2026-06-30 10:00:26,500][starVLA.training.train_starvla][INFO] - Step 1857, Loss: {'train/loss': 4.934926255373284e-05, 'train/grad_norm_pre_clip': 3.0660147666931152, 'timing/data': 0.0002928217872977257, 'timing/model': 1.229945088038221, 'learning_rate/action_model': 5.205858205354186e-06, 'learning_rate/action_query_proj': 5.205858205354186e-06, 'learning_rate/base': 1.3014645513385466e-06, 'epoch': 0.8}) +[2026-06-30 10:00:59,150][starVLA.training.train_starvla][INFO] - Step 1858, Loss: {'train/loss': 0.02102225087583065, 'train/grad_norm_pre_clip': 3.9063174724578857, 'timing/data': 0.00020147999748587608, 'timing/model': 1.2320508901029825, 'learning_rate/action_model': 5.189121764357277e-06, 'learning_rate/action_query_proj': 5.189121764357277e-06, 'learning_rate/base': 1.2972804410893194e-06, 'epoch': 0.8}) +[2026-06-30 10:01:31,806][starVLA.training.train_starvla][INFO] - Step 1859, Loss: {'train/loss': 0.0840560793876648, 'train/grad_norm_pre_clip': 2.6356210708618164, 'timing/data': 0.00029063806869089603, 'timing/model': 1.2307728598825634, 'learning_rate/action_model': 5.1725008245490845e-06, 'learning_rate/action_query_proj': 5.1725008245490845e-06, 'learning_rate/base': 1.2931252061372711e-06, 'epoch': 0.8}) +[2026-06-30 10:02:04,485][starVLA.training.train_starvla][INFO] - Step 1860, Loss: {'train/loss': 0.052722688764333725, 'train/grad_norm_pre_clip': 1.8392683267593384, 'timing/data': 0.0002050760667771101, 'timing/model': 1.230967367067933, 'learning_rate/action_model': 5.155995426940123e-06, 'learning_rate/action_query_proj': 5.155995426940123e-06, 'learning_rate/base': 1.2889988567350308e-06, 'epoch': 0.8}) +[2026-06-30 10:02:37,178][starVLA.training.train_starvla][INFO] - Step 1861, Loss: {'train/loss': 0.023647207766771317, 'train/grad_norm_pre_clip': 2.25483775138855, 'timing/data': 0.0003045538906008005, 'timing/model': 1.2317449138499796, 'learning_rate/action_model': 5.139605612255827e-06, 'learning_rate/action_query_proj': 5.139605612255827e-06, 'learning_rate/base': 1.2849014030639568e-06, 'epoch': 0.8}) +[2026-06-30 10:03:09,859][starVLA.training.train_starvla][INFO] - Step 1862, Loss: {'train/loss': 0.009061848744750023, 'train/grad_norm_pre_clip': 2.9534974098205566, 'timing/data': 0.00020264205522835255, 'timing/model': 1.2294168539810926, 'learning_rate/action_model': 5.12333142093642e-06, 'learning_rate/action_query_proj': 5.12333142093642e-06, 'learning_rate/base': 1.280832855234105e-06, 'epoch': 0.8}) +[2026-06-30 10:03:42,522][starVLA.training.train_starvla][INFO] - Step 1863, Loss: {'train/loss': 0.007063582539558411, 'train/grad_norm_pre_clip': 1.3681042194366455, 'timing/data': 0.00026334612630307674, 'timing/model': 1.2363009080290794, 'learning_rate/action_model': 5.107172893136862e-06, 'learning_rate/action_query_proj': 5.107172893136862e-06, 'learning_rate/base': 1.2767932232842154e-06, 'epoch': 0.8}) +[2026-06-30 10:04:15,199][starVLA.training.train_starvla][INFO] - Step 1864, Loss: {'train/loss': 1.5556495782220736e-05, 'train/grad_norm_pre_clip': 4.1272430419921875, 'timing/data': 0.00037008803337812424, 'timing/model': 1.228175678057596, 'learning_rate/action_model': 5.091130068726712e-06, 'learning_rate/action_query_proj': 5.091130068726712e-06, 'learning_rate/base': 1.272782517181678e-06, 'epoch': 0.8}) +[2026-06-30 10:04:47,875][starVLA.training.train_starvla][INFO] - Step 1865, Loss: {'train/loss': 2.2649737729807384e-06, 'train/grad_norm_pre_clip': 2.6241917610168457, 'timing/data': 0.0003749669995158911, 'timing/model': 1.2318827270064503, 'learning_rate/action_model': 5.075202987290051e-06, 'learning_rate/action_query_proj': 5.075202987290051e-06, 'learning_rate/base': 1.2688007468225128e-06, 'epoch': 0.8}) +[2026-06-30 10:05:20,528][starVLA.training.train_starvla][INFO] - Step 1866, Loss: {'train/loss': 2.0890824089292437e-05, 'train/grad_norm_pre_clip': 3.5439376831054688, 'timing/data': 0.00020448584109544754, 'timing/model': 1.231367923785001, 'learning_rate/action_model': 5.059391688125352e-06, 'learning_rate/action_query_proj': 5.059391688125352e-06, 'learning_rate/base': 1.264847922031338e-06, 'epoch': 0.8}) +[2026-06-30 10:05:53,165][starVLA.training.train_starvla][INFO] - Step 1867, Loss: {'train/loss': 0.14730945229530334, 'train/grad_norm_pre_clip': 3.257929563522339, 'timing/data': 0.0001950068399310112, 'timing/model': 1.2295878629665822, 'learning_rate/action_model': 5.043696210245439e-06, 'learning_rate/action_query_proj': 5.043696210245439e-06, 'learning_rate/base': 1.2609240525613597e-06, 'epoch': 0.8}) +[2026-06-30 10:06:25,990][starVLA.training.train_starvla][INFO] - Step 1868, Loss: {'train/loss': 0.0031845043413341045, 'train/grad_norm_pre_clip': 2.201946258544922, 'timing/data': 0.00020160991698503494, 'timing/model': 1.2259766180068254, 'learning_rate/action_model': 5.028116592377341e-06, 'learning_rate/action_query_proj': 5.028116592377341e-06, 'learning_rate/base': 1.2570291480943352e-06, 'epoch': 0.8}) +[2026-06-30 10:06:58,654][starVLA.training.train_starvla][INFO] - Step 1869, Loss: {'train/loss': 0.14943952858448029, 'train/grad_norm_pre_clip': 1.9837795495986938, 'timing/data': 0.0002583470195531845, 'timing/model': 1.227820001076907, 'learning_rate/action_model': 5.0126528729622095e-06, 'learning_rate/action_query_proj': 5.0126528729622095e-06, 'learning_rate/base': 1.2531632182405524e-06, 'epoch': 0.8}) +[2026-06-30 10:07:31,277][starVLA.training.train_starvla][INFO] - Step 1870, Loss: {'train/loss': 0.023475488647818565, 'train/grad_norm_pre_clip': 1.009339451789856, 'timing/data': 0.0002610129304230213, 'timing/model': 1.2318031389731914, 'learning_rate/action_model': 4.997305090155238e-06, 'learning_rate/action_query_proj': 4.997305090155238e-06, 'learning_rate/base': 1.2493262725388096e-06, 'epoch': 0.8}) +[2026-06-30 10:08:03,926][starVLA.training.train_starvla][INFO] - Step 1871, Loss: {'train/loss': 0.0503443218767643, 'train/grad_norm_pre_clip': 1.654233694076538, 'timing/data': 0.00022139702923595905, 'timing/model': 1.2293467710260302, 'learning_rate/action_model': 4.982073281825563e-06, 'learning_rate/action_query_proj': 4.982073281825563e-06, 'learning_rate/base': 1.2455183204563908e-06, 'epoch': 0.81}) +[2026-06-30 10:08:36,593][starVLA.training.train_starvla][INFO] - Step 1872, Loss: {'train/loss': 0.17414149641990662, 'train/grad_norm_pre_clip': 3.134537935256958, 'timing/data': 0.00031098583713173866, 'timing/model': 1.2310007971245795, 'learning_rate/action_model': 4.966957485556151e-06, 'learning_rate/action_query_proj': 4.966957485556151e-06, 'learning_rate/base': 1.2417393713890377e-06, 'epoch': 0.81}) +[2026-06-30 10:09:09,267][starVLA.training.train_starvla][INFO] - Step 1873, Loss: {'train/loss': 0.05238217115402222, 'train/grad_norm_pre_clip': 1.6952770948410034, 'timing/data': 0.00025316793471574783, 'timing/model': 1.2299592059571296, 'learning_rate/action_model': 4.9519577386437255e-06, 'learning_rate/action_query_proj': 4.9519577386437255e-06, 'learning_rate/base': 1.2379894346609314e-06, 'epoch': 0.81}) +[2026-06-30 10:09:41,828][starVLA.training.train_starvla][INFO] - Step 1874, Loss: {'train/loss': 0.275452584028244, 'train/grad_norm_pre_clip': 1.7155829668045044, 'timing/data': 0.00019659101963043213, 'timing/model': 1.2306871030014008, 'learning_rate/action_model': 4.937074078098667e-06, 'learning_rate/action_query_proj': 4.937074078098667e-06, 'learning_rate/base': 1.2342685195246668e-06, 'epoch': 0.81}) +[2026-06-30 10:10:14,468][starVLA.training.train_starvla][INFO] - Step 1875, Loss: {'train/loss': 0.0006625018431805074, 'train/grad_norm_pre_clip': 2.068406820297241, 'timing/data': 0.00021034711971879005, 'timing/model': 1.2330662850290537, 'learning_rate/action_model': 4.922306540644938e-06, 'learning_rate/action_query_proj': 4.922306540644938e-06, 'learning_rate/base': 1.2305766351612346e-06, 'epoch': 0.81}) +[2026-06-30 10:10:47,154][starVLA.training.train_starvla][INFO] - Step 1876, Loss: {'train/loss': 0.05039610713720322, 'train/grad_norm_pre_clip': 2.1400399208068848, 'timing/data': 0.0002103070728480816, 'timing/model': 1.2292478971648961, 'learning_rate/action_model': 4.907655162719969e-06, 'learning_rate/action_query_proj': 4.907655162719969e-06, 'learning_rate/base': 1.2269137906799922e-06, 'epoch': 0.81}) +[2026-06-30 10:11:19,811][starVLA.training.train_starvla][INFO] - Step 1877, Loss: {'train/loss': 0.31492093205451965, 'train/grad_norm_pre_clip': 1.8818254470825195, 'timing/data': 0.00019682105630636215, 'timing/model': 1.235262834932655, 'learning_rate/action_model': 4.893119980474562e-06, 'learning_rate/action_query_proj': 4.893119980474562e-06, 'learning_rate/base': 1.2232799951186405e-06, 'epoch': 0.81}) +[2026-06-30 10:11:52,520][starVLA.training.train_starvla][INFO] - Step 1878, Loss: {'train/loss': 0.0014665927737951279, 'train/grad_norm_pre_clip': 2.4053962230682373, 'timing/data': 0.0002031838521361351, 'timing/model': 1.2324366378597915, 'learning_rate/action_model': 4.878701029772857e-06, 'learning_rate/action_query_proj': 4.878701029772857e-06, 'learning_rate/base': 1.2196752574432143e-06, 'epoch': 0.81}) +[2026-06-30 10:12:25,182][starVLA.training.train_starvla][INFO] - Step 1879, Loss: {'train/loss': 0.41807910799980164, 'train/grad_norm_pre_clip': 1.9629212617874146, 'timing/data': 0.00019889511168003082, 'timing/model': 1.2306345719844103, 'learning_rate/action_model': 4.864398346192166e-06, 'learning_rate/action_query_proj': 4.864398346192166e-06, 'learning_rate/base': 1.2160995865480414e-06, 'epoch': 0.81}) +[2026-06-30 10:12:57,840][starVLA.training.train_starvla][INFO] - Step 1880, Loss: {'train/loss': 0.00020341831259429455, 'train/grad_norm_pre_clip': 2.7335715293884277, 'timing/data': 0.00019922596402466297, 'timing/model': 1.2299157888628542, 'learning_rate/action_model': 4.850211965022947e-06, 'learning_rate/action_query_proj': 4.850211965022947e-06, 'learning_rate/base': 1.2125529912557367e-06, 'epoch': 0.81}) +[2026-06-30 10:13:30,489][starVLA.training.train_starvla][INFO] - Step 1881, Loss: {'train/loss': 2.7120076993014663e-06, 'train/grad_norm_pre_clip': 2.417166233062744, 'timing/data': 0.00029697990976274014, 'timing/model': 1.2311440578196198, 'learning_rate/action_model': 4.836141921268676e-06, 'learning_rate/action_query_proj': 4.836141921268676e-06, 'learning_rate/base': 1.209035480317169e-06, 'epoch': 0.81}) +[2026-06-30 10:14:03,180][starVLA.training.train_starvla][INFO] - Step 1882, Loss: {'train/loss': 0.001210344024002552, 'train/grad_norm_pre_clip': 3.315019369125366, 'timing/data': 0.00022140704095363617, 'timing/model': 1.2300091010984033, 'learning_rate/action_model': 4.822188249645799e-06, 'learning_rate/action_query_proj': 4.822188249645799e-06, 'learning_rate/base': 1.2055470624114498e-06, 'epoch': 0.81}) +[2026-06-30 10:14:35,850][starVLA.training.train_starvla][INFO] - Step 1883, Loss: {'train/loss': 0.02531340904533863, 'train/grad_norm_pre_clip': 2.615644931793213, 'timing/data': 0.00020160013809800148, 'timing/model': 1.2309612799435854, 'learning_rate/action_model': 4.808350984583606e-06, 'learning_rate/action_query_proj': 4.808350984583606e-06, 'learning_rate/base': 1.2020877461459015e-06, 'epoch': 0.81}) +[2026-06-30 10:15:08,523][starVLA.training.train_starvla][INFO] - Step 1884, Loss: {'train/loss': 0.028249934315681458, 'train/grad_norm_pre_clip': 2.1964385509490967, 'timing/data': 0.00020359409973025322, 'timing/model': 1.2264838861301541, 'learning_rate/action_model': 4.794630160224178e-06, 'learning_rate/action_query_proj': 4.794630160224178e-06, 'learning_rate/base': 1.1986575400560446e-06, 'epoch': 0.81}) +[2026-06-30 10:15:41,250][starVLA.training.train_starvla][INFO] - Step 1885, Loss: {'train/loss': 0.0006550948019139469, 'train/grad_norm_pre_clip': 1.5668638944625854, 'timing/data': 0.00025963992811739445, 'timing/model': 1.2314872660208493, 'learning_rate/action_model': 4.781025810422275e-06, 'learning_rate/action_query_proj': 4.781025810422275e-06, 'learning_rate/base': 1.1952564526055686e-06, 'epoch': 0.81}) +[2026-06-30 10:16:13,911][starVLA.training.train_starvla][INFO] - Step 1886, Loss: {'train/loss': 0.021099373698234558, 'train/grad_norm_pre_clip': 3.649562120437622, 'timing/data': 0.00024414993822574615, 'timing/model': 1.2322194080334157, 'learning_rate/action_model': 4.767537968745294e-06, 'learning_rate/action_query_proj': 4.767537968745294e-06, 'learning_rate/base': 1.1918844921863236e-06, 'epoch': 0.81}) +[2026-06-30 10:16:46,610][starVLA.training.train_starvla][INFO] - Step 1887, Loss: {'train/loss': 6.284814298851416e-05, 'train/grad_norm_pre_clip': 2.769683837890625, 'timing/data': 0.00019429600797593594, 'timing/model': 1.230412453180179, 'learning_rate/action_model': 4.754166668473129e-06, 'learning_rate/action_query_proj': 4.754166668473129e-06, 'learning_rate/base': 1.1885416671182822e-06, 'epoch': 0.81}) +[2026-06-30 10:17:19,266][starVLA.training.train_starvla][INFO] - Step 1888, Loss: {'train/loss': 0.015916330739855766, 'train/grad_norm_pre_clip': 2.278001070022583, 'timing/data': 0.0002623640466481447, 'timing/model': 1.2260576400440186, 'learning_rate/action_model': 4.740911942598144e-06, 'learning_rate/action_query_proj': 4.740911942598144e-06, 'learning_rate/base': 1.185227985649536e-06, 'epoch': 0.81}) +[2026-06-30 10:17:51,927][starVLA.training.train_starvla][INFO] - Step 1889, Loss: {'train/loss': 0.002769127255305648, 'train/grad_norm_pre_clip': 2.3005945682525635, 'timing/data': 0.00020670006051659584, 'timing/model': 1.2337970901280642, 'learning_rate/action_model': 4.727773823825048e-06, 'learning_rate/action_query_proj': 4.727773823825048e-06, 'learning_rate/base': 1.181943455956262e-06, 'epoch': 0.81}) +[2026-06-30 10:18:24,618][starVLA.training.train_starvla][INFO] - Step 1890, Loss: {'train/loss': 0.006076221354305744, 'train/grad_norm_pre_clip': 4.182196140289307, 'timing/data': 0.00027066003531217575, 'timing/model': 1.2346448609605432, 'learning_rate/action_model': 4.714752344570854e-06, 'learning_rate/action_query_proj': 4.714752344570854e-06, 'learning_rate/base': 1.1786880861427134e-06, 'epoch': 0.81}) +[2026-06-30 10:18:57,293][starVLA.training.train_starvla][INFO] - Step 1891, Loss: {'train/loss': 0.0028561085928231478, 'train/grad_norm_pre_clip': 1.995593547821045, 'timing/data': 0.00020384392701089382, 'timing/model': 1.2397701188456267, 'learning_rate/action_model': 4.701847536964758e-06, 'learning_rate/action_query_proj': 4.701847536964758e-06, 'learning_rate/base': 1.1754618842411895e-06, 'epoch': 0.81}) +[2026-06-30 10:19:29,882][starVLA.training.train_starvla][INFO] - Step 1892, Loss: {'train/loss': 0.0002375611220486462, 'train/grad_norm_pre_clip': 2.660512924194336, 'timing/data': 0.0002743271179497242, 'timing/model': 1.2234336959663779, 'learning_rate/action_model': 4.689059432848093e-06, 'learning_rate/action_query_proj': 4.689059432848093e-06, 'learning_rate/base': 1.1722648582120233e-06, 'epoch': 0.81}) +[2026-06-30 10:20:02,494][starVLA.training.train_starvla][INFO] - Step 1893, Loss: {'train/loss': 9.953876542567741e-06, 'train/grad_norm_pre_clip': 1.9951337575912476, 'timing/data': 0.000244801864027977, 'timing/model': 1.2318045368883759, 'learning_rate/action_model': 4.67638806377424e-06, 'learning_rate/action_query_proj': 4.67638806377424e-06, 'learning_rate/base': 1.16909701594356e-06, 'epoch': 0.81}) +[2026-06-30 10:20:35,149][starVLA.training.train_starvla][INFO] - Step 1894, Loss: {'train/loss': 0.00019659067038446665, 'train/grad_norm_pre_clip': 1.9795238971710205, 'timing/data': 0.0001987649593502283, 'timing/model': 1.229053518967703, 'learning_rate/action_model': 4.66383346100854e-06, 'learning_rate/action_query_proj': 4.66383346100854e-06, 'learning_rate/base': 1.165958365252135e-06, 'epoch': 0.81}) +[2026-06-30 10:21:07,843][starVLA.training.train_starvla][INFO] - Step 1895, Loss: {'train/loss': 4.136368806939572e-05, 'train/grad_norm_pre_clip': 2.464768171310425, 'timing/data': 0.0002487890888005495, 'timing/model': 1.2366275968961418, 'learning_rate/action_model': 4.651395655528218e-06, 'learning_rate/action_query_proj': 4.651395655528218e-06, 'learning_rate/base': 1.1628489138820546e-06, 'epoch': 0.82}) +[2026-06-30 10:21:40,504][starVLA.training.train_starvla][INFO] - Step 1896, Loss: {'train/loss': 3.0398321086977376e-06, 'train/grad_norm_pre_clip': 2.320237159729004, 'timing/data': 0.0002701289486140013, 'timing/model': 1.218788119032979, 'learning_rate/action_model': 4.639074678022334e-06, 'learning_rate/action_query_proj': 4.639074678022334e-06, 'learning_rate/base': 1.1597686695055834e-06, 'epoch': 0.82}) +[2026-06-30 10:22:13,173][starVLA.training.train_starvla][INFO] - Step 1897, Loss: {'train/loss': 0.46816009283065796, 'train/grad_norm_pre_clip': 2.9149365425109863, 'timing/data': 0.00031112716533243656, 'timing/model': 1.229568504029885, 'learning_rate/action_model': 4.6268705588916706e-06, 'learning_rate/action_query_proj': 4.6268705588916706e-06, 'learning_rate/base': 1.1567176397229176e-06, 'epoch': 0.82}) +[2026-06-30 10:22:45,815][starVLA.training.train_starvla][INFO] - Step 1898, Loss: {'train/loss': 0.07629947364330292, 'train/grad_norm_pre_clip': 2.879753589630127, 'timing/data': 0.00024413014762103558, 'timing/model': 1.2328362080734223, 'learning_rate/action_model': 4.61478332824868e-06, 'learning_rate/action_query_proj': 4.61478332824868e-06, 'learning_rate/base': 1.15369583206217e-06, 'epoch': 0.82}) +[2026-06-30 10:23:18,509][starVLA.training.train_starvla][INFO] - Step 1899, Loss: {'train/loss': 0.013924297876656055, 'train/grad_norm_pre_clip': 2.373065948486328, 'timing/data': 0.00021470501087605953, 'timing/model': 1.2270844399463385, 'learning_rate/action_model': 4.6028130159174e-06, 'learning_rate/action_query_proj': 4.6028130159174e-06, 'learning_rate/base': 1.15070325397935e-06, 'epoch': 0.82}) +[2026-06-30 10:23:51,221][starVLA.training.train_starvla][INFO] - Step 1900, Loss: {'train/loss': 0.0005572257214225829, 'train/grad_norm_pre_clip': 1.7954447269439697, 'timing/data': 0.00033312803134322166, 'timing/model': 1.2301667449064553, 'learning_rate/action_model': 4.590959651433392e-06, 'learning_rate/action_query_proj': 4.590959651433392e-06, 'learning_rate/base': 1.147739912858348e-06, 'epoch': 0.82}) +[2026-06-30 10:24:23,856][starVLA.training.train_starvla][INFO] - Step 1901, Loss: {'train/loss': 0.007998466491699219, 'train/grad_norm_pre_clip': 1.9222025871276855, 'timing/data': 0.0002009989693760872, 'timing/model': 1.229396483162418, 'learning_rate/action_model': 4.579223264043646e-06, 'learning_rate/action_query_proj': 4.579223264043646e-06, 'learning_rate/base': 1.1448058160109114e-06, 'epoch': 0.82}) +[2026-06-30 10:24:56,515][starVLA.training.train_starvla][INFO] - Step 1902, Loss: {'train/loss': 0.014566606841981411, 'train/grad_norm_pre_clip': 3.8442554473876953, 'timing/data': 0.0002900769468396902, 'timing/model': 1.2309758530464023, 'learning_rate/action_model': 4.5676038827065386e-06, 'learning_rate/action_query_proj': 4.5676038827065386e-06, 'learning_rate/base': 1.1419009706766346e-06, 'epoch': 0.82}) +[2026-06-30 10:25:29,191][starVLA.training.train_starvla][INFO] - Step 1903, Loss: {'train/loss': 7.599550372106023e-06, 'train/grad_norm_pre_clip': 3.0991010665893555, 'timing/data': 0.0002445119898766279, 'timing/model': 1.232517461059615, 'learning_rate/action_model': 4.5561015360917285e-06, 'learning_rate/action_query_proj': 4.5561015360917285e-06, 'learning_rate/base': 1.1390253840229321e-06, 'epoch': 0.82}) +[2026-06-30 10:26:01,845][starVLA.training.train_starvla][INFO] - Step 1904, Loss: {'train/loss': 0.0005122623988427222, 'train/grad_norm_pre_clip': 4.043822765350342, 'timing/data': 0.0002138940617442131, 'timing/model': 1.2287970441393554, 'learning_rate/action_model': 4.54471625258013e-06, 'learning_rate/action_query_proj': 4.54471625258013e-06, 'learning_rate/base': 1.1361790631450326e-06, 'epoch': 0.82}) +[2026-06-30 10:26:34,489][starVLA.training.train_starvla][INFO] - Step 1905, Loss: {'train/loss': 0.01074988953769207, 'train/grad_norm_pre_clip': 1.9708949327468872, 'timing/data': 0.00019619986414909363, 'timing/model': 1.2310214908793569, 'learning_rate/action_model': 4.533448060263783e-06, 'learning_rate/action_query_proj': 4.533448060263783e-06, 'learning_rate/base': 1.1333620150659457e-06, 'epoch': 0.82}) +[2026-06-30 10:27:07,104][starVLA.training.train_starvla][INFO] - Step 1906, Loss: {'train/loss': 0.0558890774846077, 'train/grad_norm_pre_clip': 1.9055287837982178, 'timing/data': 0.00029193004593253136, 'timing/model': 1.235008561052382, 'learning_rate/action_model': 4.522296986945833e-06, 'learning_rate/action_query_proj': 4.522296986945833e-06, 'learning_rate/base': 1.1305742467364583e-06, 'epoch': 0.82}) +[2026-06-30 10:27:39,760][starVLA.training.train_starvla][INFO] - Step 1907, Loss: {'train/loss': 0.8519753217697144, 'train/grad_norm_pre_clip': 3.137610912322998, 'timing/data': 0.0004181589465588331, 'timing/model': 1.2339441839139909, 'learning_rate/action_model': 4.51126306014045e-06, 'learning_rate/action_query_proj': 4.51126306014045e-06, 'learning_rate/base': 1.1278157650351124e-06, 'epoch': 0.82}) +[2026-06-30 10:28:12,443][starVLA.training.train_starvla][INFO] - Step 1908, Loss: {'train/loss': 0.06816208362579346, 'train/grad_norm_pre_clip': 2.003460645675659, 'timing/data': 0.00019423593766987324, 'timing/model': 1.2304940270259976, 'learning_rate/action_model': 4.5003463070727545e-06, 'learning_rate/action_query_proj': 4.5003463070727545e-06, 'learning_rate/base': 1.1250865767681886e-06, 'epoch': 0.82}) +[2026-06-30 10:28:45,096][starVLA.training.train_starvla][INFO] - Step 1909, Loss: {'train/loss': 0.0001848414249252528, 'train/grad_norm_pre_clip': 2.008640766143799, 'timing/data': 0.000289025017991662, 'timing/model': 1.23141877213493, 'learning_rate/action_model': 4.4895467546787435e-06, 'learning_rate/action_query_proj': 4.4895467546787435e-06, 'learning_rate/base': 1.1223866886696859e-06, 'epoch': 0.82}) +[2026-06-30 10:29:17,760][starVLA.training.train_starvla][INFO] - Step 1910, Loss: {'train/loss': 0.006354151293635368, 'train/grad_norm_pre_clip': 1.6112704277038574, 'timing/data': 0.00020269211381673813, 'timing/model': 1.2300779030192643, 'learning_rate/action_model': 4.478864429605238e-06, 'learning_rate/action_query_proj': 4.478864429605238e-06, 'learning_rate/base': 1.1197161074013095e-06, 'epoch': 0.82}) +[2026-06-30 10:29:50,411][starVLA.training.train_starvla][INFO] - Step 1911, Loss: {'train/loss': 1.123537185776513e-05, 'train/grad_norm_pre_clip': 2.2113752365112305, 'timing/data': 0.00020449585281312466, 'timing/model': 1.2321211178787053, 'learning_rate/action_model': 4.4682993582098165e-06, 'learning_rate/action_query_proj': 4.4682993582098165e-06, 'learning_rate/base': 1.1170748395524541e-06, 'epoch': 0.82}) +[2026-06-30 10:30:23,051][starVLA.training.train_starvla][INFO] - Step 1912, Loss: {'train/loss': 0.12115447968244553, 'train/grad_norm_pre_clip': 1.8146687746047974, 'timing/data': 0.00019927602261304855, 'timing/model': 1.2289680810645223, 'learning_rate/action_model': 4.4578515665607426e-06, 'learning_rate/action_query_proj': 4.4578515665607426e-06, 'learning_rate/base': 1.1144628916401856e-06, 'epoch': 0.82}) +[2026-06-30 10:30:55,707][starVLA.training.train_starvla][INFO] - Step 1913, Loss: {'train/loss': 5.662413968821056e-06, 'train/grad_norm_pre_clip': 1.3113669157028198, 'timing/data': 0.00020451610907912254, 'timing/model': 1.2335288440808654, 'learning_rate/action_model': 4.447521080436903e-06, 'learning_rate/action_query_proj': 4.447521080436903e-06, 'learning_rate/base': 1.1118802701092259e-06, 'epoch': 0.82}) +[2026-06-30 10:31:28,397][starVLA.training.train_starvla][INFO] - Step 1914, Loss: {'train/loss': 0.0062558287754654884, 'train/grad_norm_pre_clip': 2.7233080863952637, 'timing/data': 0.0002658411394804716, 'timing/model': 1.2357524279505014, 'learning_rate/action_model': 4.437307925327743e-06, 'learning_rate/action_query_proj': 4.437307925327743e-06, 'learning_rate/base': 1.1093269813319357e-06, 'epoch': 0.82}) +[2026-06-30 10:32:01,072][starVLA.training.train_starvla][INFO] - Step 1915, Loss: {'train/loss': 0.14671412110328674, 'train/grad_norm_pre_clip': 2.574791431427002, 'timing/data': 0.0002797669731080532, 'timing/model': 1.21884472691454, 'learning_rate/action_model': 4.427212126433212e-06, 'learning_rate/action_query_proj': 4.427212126433212e-06, 'learning_rate/base': 1.106803031608303e-06, 'epoch': 0.82}) +[2026-06-30 10:32:33,766][starVLA.training.train_starvla][INFO] - Step 1916, Loss: {'train/loss': 0.01793311908841133, 'train/grad_norm_pre_clip': 2.665872097015381, 'timing/data': 0.00027756416238844395, 'timing/model': 1.232940445188433, 'learning_rate/action_model': 4.417233708663689e-06, 'learning_rate/action_query_proj': 4.417233708663689e-06, 'learning_rate/base': 1.1043084271659222e-06, 'epoch': 0.82}) +[2026-06-30 10:33:06,447][starVLA.training.train_starvla][INFO] - Step 1917, Loss: {'train/loss': 0.007384856231510639, 'train/grad_norm_pre_clip': 1.2915548086166382, 'timing/data': 0.0003629438579082489, 'timing/model': 1.2270297659561038, 'learning_rate/action_model': 4.407372696639919e-06, 'learning_rate/action_query_proj': 4.407372696639919e-06, 'learning_rate/base': 1.1018431741599799e-06, 'epoch': 0.82}) +[2026-06-30 10:33:39,140][starVLA.training.train_starvla][INFO] - Step 1918, Loss: {'train/loss': 0.0037364743184298277, 'train/grad_norm_pre_clip': 2.7044994831085205, 'timing/data': 0.00020934408530592918, 'timing/model': 1.2344437511637807, 'learning_rate/action_model': 4.397629114692985e-06, 'learning_rate/action_query_proj': 4.397629114692985e-06, 'learning_rate/base': 1.0994072786732463e-06, 'epoch': 0.83}) +[2026-06-30 10:34:11,886][starVLA.training.train_starvla][INFO] - Step 1919, Loss: {'train/loss': 0.22784669697284698, 'train/grad_norm_pre_clip': 3.5266151428222656, 'timing/data': 0.00028039910830557346, 'timing/model': 1.2288548490032554, 'learning_rate/action_model': 4.388002986864195e-06, 'learning_rate/action_query_proj': 4.388002986864195e-06, 'learning_rate/base': 1.0970007467160488e-06, 'epoch': 0.83}) +[2026-06-30 10:34:44,563][starVLA.training.train_starvla][INFO] - Step 1920, Loss: {'train/loss': 4.371674731373787e-05, 'train/grad_norm_pre_clip': 1.8997794389724731, 'timing/data': 0.0002620450686663389, 'timing/model': 1.2365913020912558, 'learning_rate/action_model': 4.378494336905067e-06, 'learning_rate/action_query_proj': 4.378494336905067e-06, 'learning_rate/base': 1.0946235842262667e-06, 'epoch': 0.83}) +[2026-06-30 10:35:17,442][starVLA.training.train_starvla][INFO] - Step 1921, Loss: {'train/loss': 0.020721156150102615, 'train/grad_norm_pre_clip': 2.5271522998809814, 'timing/data': 0.00024824892170727253, 'timing/model': 1.2315865620039403, 'learning_rate/action_model': 4.3691031882772436e-06, 'learning_rate/action_query_proj': 4.3691031882772436e-06, 'learning_rate/base': 1.0922757970693109e-06, 'epoch': 0.83}) +[2026-06-30 10:35:50,092][starVLA.training.train_starvla][INFO] - Step 1922, Loss: {'train/loss': 0.2770223319530487, 'train/grad_norm_pre_clip': 1.604909896850586, 'timing/data': 0.00021197902970016003, 'timing/model': 1.2315157889388502, 'learning_rate/action_model': 4.359829564152461e-06, 'learning_rate/action_query_proj': 4.359829564152461e-06, 'learning_rate/base': 1.0899573910381153e-06, 'epoch': 0.83}) +[2026-06-30 10:36:22,830][starVLA.training.train_starvla][INFO] - Step 1923, Loss: {'train/loss': 0.00175731279887259, 'train/grad_norm_pre_clip': 4.655633926391602, 'timing/data': 0.00029559689573943615, 'timing/model': 1.2315753020811826, 'learning_rate/action_model': 4.350673487412458e-06, 'learning_rate/action_query_proj': 4.350673487412458e-06, 'learning_rate/base': 1.0876683718531146e-06, 'epoch': 0.83}) +[2026-06-30 10:36:55,470][starVLA.training.train_starvla][INFO] - Step 1924, Loss: {'train/loss': 0.4366435408592224, 'train/grad_norm_pre_clip': 3.124265193939209, 'timing/data': 0.00025772606022655964, 'timing/model': 1.229341738857329, 'learning_rate/action_model': 4.341634980648948e-06, 'learning_rate/action_query_proj': 4.341634980648948e-06, 'learning_rate/base': 1.085408745162237e-06, 'epoch': 0.83}) +[2026-06-30 10:37:28,157][starVLA.training.train_starvla][INFO] - Step 1925, Loss: {'train/loss': 1.415595761500299e-05, 'train/grad_norm_pre_clip': 1.0883283615112305, 'timing/data': 0.00020171096548438072, 'timing/model': 1.2323289432097226, 'learning_rate/action_model': 4.3327140661635375e-06, 'learning_rate/action_query_proj': 4.3327140661635375e-06, 'learning_rate/base': 1.0831785165408844e-06, 'epoch': 0.83}) +[2026-06-30 10:38:00,840][starVLA.training.train_starvla][INFO] - Step 1926, Loss: {'train/loss': 0.00015159280155785382, 'train/grad_norm_pre_clip': 1.6744985580444336, 'timing/data': 0.00020531797781586647, 'timing/model': 1.233039986807853, 'learning_rate/action_model': 4.32391076596771e-06, 'learning_rate/action_query_proj': 4.32391076596771e-06, 'learning_rate/base': 1.0809776914919275e-06, 'epoch': 0.83}) +[2026-06-30 10:38:33,548][starVLA.training.train_starvla][INFO] - Step 1927, Loss: {'train/loss': 0.11982021480798721, 'train/grad_norm_pre_clip': 1.7713507413864136, 'timing/data': 0.00020005716942250729, 'timing/model': 1.2273699149955064, 'learning_rate/action_model': 4.315225101782724e-06, 'learning_rate/action_query_proj': 4.315225101782724e-06, 'learning_rate/base': 1.078806275445681e-06, 'epoch': 0.83}) +[2026-06-30 10:39:06,228][starVLA.training.train_starvla][INFO] - Step 1928, Loss: {'train/loss': 0.062155865132808685, 'train/grad_norm_pre_clip': 3.002919912338257, 'timing/data': 0.0003091939724981785, 'timing/model': 1.2305684830062091, 'learning_rate/action_model': 4.306657095039597e-06, 'learning_rate/action_query_proj': 4.306657095039597e-06, 'learning_rate/base': 1.0766642737598992e-06, 'epoch': 0.83}) +[2026-06-30 10:39:38,918][starVLA.training.train_starvla][INFO] - Step 1929, Loss: {'train/loss': 0.060031477361917496, 'train/grad_norm_pre_clip': 4.052823066711426, 'timing/data': 0.0002623549662530422, 'timing/model': 1.2335451729595661, 'learning_rate/action_model': 4.298206766879033e-06, 'learning_rate/action_query_proj': 4.298206766879033e-06, 'learning_rate/base': 1.0745516917197582e-06, 'epoch': 0.83}) +[2026-06-30 10:40:11,611][starVLA.training.train_starvla][INFO] - Step 1930, Loss: {'train/loss': 0.00031791991204954684, 'train/grad_norm_pre_clip': 2.453427314758301, 'timing/data': 0.0003921289462596178, 'timing/model': 1.234728811075911, 'learning_rate/action_model': 4.289874138151379e-06, 'learning_rate/action_query_proj': 4.289874138151379e-06, 'learning_rate/base': 1.0724685345378448e-06, 'epoch': 0.83}) +[2026-06-30 10:40:44,267][starVLA.training.train_starvla][INFO] - Step 1931, Loss: {'train/loss': 0.03005373477935791, 'train/grad_norm_pre_clip': 2.5511648654937744, 'timing/data': 0.0002486391458660364, 'timing/model': 1.2298499529715627, 'learning_rate/action_model': 4.281659229416556e-06, 'learning_rate/action_query_proj': 4.281659229416556e-06, 'learning_rate/base': 1.070414807354139e-06, 'epoch': 0.83}) +[2026-06-30 10:41:16,952][starVLA.training.train_starvla][INFO] - Step 1932, Loss: {'train/loss': 0.0009611094719730318, 'train/grad_norm_pre_clip': 2.6394054889678955, 'timing/data': 0.00023191701620817184, 'timing/model': 1.2281056181527674, 'learning_rate/action_model': 4.273562060944051e-06, 'learning_rate/action_query_proj': 4.273562060944051e-06, 'learning_rate/base': 1.0683905152360128e-06, 'epoch': 0.83}) +[2026-06-30 10:41:49,584][starVLA.training.train_starvla][INFO] - Step 1933, Loss: {'train/loss': 0.000455672008683905, 'train/grad_norm_pre_clip': 1.8893928527832031, 'timing/data': 0.000201069051399827, 'timing/model': 1.2334141281899065, 'learning_rate/action_model': 4.265582652712817e-06, 'learning_rate/action_query_proj': 4.265582652712817e-06, 'learning_rate/base': 1.0663956631782042e-06, 'epoch': 0.83}) +[2026-06-30 10:42:22,242][starVLA.training.train_starvla][INFO] - Step 1934, Loss: {'train/loss': 7.2717339207883924e-06, 'train/grad_norm_pre_clip': 1.3100018501281738, 'timing/data': 0.00019814400002360344, 'timing/model': 1.2299534149933606, 'learning_rate/action_model': 4.257721024411248e-06, 'learning_rate/action_query_proj': 4.257721024411248e-06, 'learning_rate/base': 1.064430256102812e-06, 'epoch': 0.83}) +[2026-06-30 10:42:54,921][starVLA.training.train_starvla][INFO] - Step 1935, Loss: {'train/loss': 0.08750951290130615, 'train/grad_norm_pre_clip': 2.0861923694610596, 'timing/data': 0.00019599893130362034, 'timing/model': 1.2339996239170432, 'learning_rate/action_model': 4.249977195437129e-06, 'learning_rate/action_query_proj': 4.249977195437129e-06, 'learning_rate/base': 1.0624942988592822e-06, 'epoch': 0.83}) +[2026-06-30 10:43:27,608][starVLA.training.train_starvla][INFO] - Step 1936, Loss: {'train/loss': 0.0006689733127132058, 'train/grad_norm_pre_clip': 1.553223967552185, 'timing/data': 0.00028567900881171227, 'timing/model': 1.2321081629488617, 'learning_rate/action_model': 4.242351184897591e-06, 'learning_rate/action_query_proj': 4.242351184897591e-06, 'learning_rate/base': 1.0605877962243977e-06, 'epoch': 0.83}) +[2026-06-30 10:44:00,157][starVLA.training.train_starvla][INFO] - Step 1937, Loss: {'train/loss': 0.08226016908884048, 'train/grad_norm_pre_clip': 1.8127819299697876, 'timing/data': 0.0002618241123855114, 'timing/model': 1.2267901089508086, 'learning_rate/action_model': 4.2348430116090615e-06, 'learning_rate/action_query_proj': 4.2348430116090615e-06, 'learning_rate/base': 1.0587107529022654e-06, 'epoch': 0.83}) +[2026-06-30 10:44:32,818][starVLA.training.train_starvla][INFO] - Step 1938, Loss: {'train/loss': 0.3573574721813202, 'train/grad_norm_pre_clip': 1.422838807106018, 'timing/data': 0.0002731147687882185, 'timing/model': 1.2297600680030882, 'learning_rate/action_model': 4.227452694097201e-06, 'learning_rate/action_query_proj': 4.227452694097201e-06, 'learning_rate/base': 1.0568631735243002e-06, 'epoch': 0.83}) +[2026-06-30 10:45:05,485][starVLA.training.train_starvla][INFO] - Step 1939, Loss: {'train/loss': 0.3027019202709198, 'train/grad_norm_pre_clip': 3.2838969230651855, 'timing/data': 0.00030246004462242126, 'timing/model': 1.2312777580227703, 'learning_rate/action_model': 4.220180250596888e-06, 'learning_rate/action_query_proj': 4.220180250596888e-06, 'learning_rate/base': 1.055045062649222e-06, 'epoch': 0.83}) +[2026-06-30 10:45:38,152][starVLA.training.train_starvla][INFO] - Step 1940, Loss: {'train/loss': 0.0002828810829669237, 'train/grad_norm_pre_clip': 2.2143383026123047, 'timing/data': 0.0001967810094356537, 'timing/model': 1.2294549869839102, 'learning_rate/action_model': 4.21302569905216e-06, 'learning_rate/action_query_proj': 4.21302569905216e-06, 'learning_rate/base': 1.05325642476304e-06, 'epoch': 0.83}) +[2026-06-30 10:46:10,809][starVLA.training.train_starvla][INFO] - Step 1941, Loss: {'train/loss': 6.493500404758379e-05, 'train/grad_norm_pre_clip': 3.1500024795532227, 'timing/data': 0.0002515639644116163, 'timing/model': 1.231094527989626, 'learning_rate/action_model': 4.205989057116154e-06, 'learning_rate/action_query_proj': 4.205989057116154e-06, 'learning_rate/base': 1.0514972642790385e-06, 'epoch': 0.84}) +[2026-06-30 10:46:43,485][starVLA.training.train_starvla][INFO] - Step 1942, Loss: {'train/loss': 0.000484159158077091, 'train/grad_norm_pre_clip': 3.939373731613159, 'timing/data': 0.0002694691065698862, 'timing/model': 1.2306626699864864, 'learning_rate/action_model': 4.199070342151088e-06, 'learning_rate/action_query_proj': 4.199070342151088e-06, 'learning_rate/base': 1.049767585537772e-06, 'epoch': 0.84}) +[2026-06-30 10:47:16,107][starVLA.training.train_starvla][INFO] - Step 1943, Loss: {'train/loss': 1.3679029507329687e-05, 'train/grad_norm_pre_clip': 2.3962273597717285, 'timing/data': 0.00019820407032966614, 'timing/model': 1.2279288119170815, 'learning_rate/action_model': 4.192269571228198e-06, 'learning_rate/action_query_proj': 4.192269571228198e-06, 'learning_rate/base': 1.0480673928070494e-06, 'epoch': 0.84}) +[2026-06-30 10:47:48,793][starVLA.training.train_starvla][INFO] - Step 1944, Loss: {'train/loss': 0.0004516594344750047, 'train/grad_norm_pre_clip': 1.719489574432373, 'timing/data': 0.00021660909987986088, 'timing/model': 1.2310995100997388, 'learning_rate/action_model': 4.18558676112772e-06, 'learning_rate/action_query_proj': 4.18558676112772e-06, 'learning_rate/base': 1.04639669028193e-06, 'epoch': 0.84}) +[2026-06-30 10:48:21,481][starVLA.training.train_starvla][INFO] - Step 1945, Loss: {'train/loss': 0.005012949928641319, 'train/grad_norm_pre_clip': 1.1467022895812988, 'timing/data': 0.0002272389829158783, 'timing/model': 1.228269309969619, 'learning_rate/action_model': 4.179021928338819e-06, 'learning_rate/action_query_proj': 4.179021928338819e-06, 'learning_rate/base': 1.0447554820847048e-06, 'epoch': 0.84}) +[2026-06-30 10:48:54,137][starVLA.training.train_starvla][INFO] - Step 1946, Loss: {'train/loss': 0.028638998046517372, 'train/grad_norm_pre_clip': 2.1231155395507812, 'timing/data': 0.00027294503524899483, 'timing/model': 1.2292085499502718, 'learning_rate/action_model': 4.1725750890595644e-06, 'learning_rate/action_query_proj': 4.1725750890595644e-06, 'learning_rate/base': 1.0431437722648911e-06, 'epoch': 0.84}) +[2026-06-30 10:49:26,770][starVLA.training.train_starvla][INFO] - Step 1947, Loss: {'train/loss': 6.109447895141784e-06, 'train/grad_norm_pre_clip': 3.807675361633301, 'timing/data': 0.0001957290805876255, 'timing/model': 1.231549713993445, 'learning_rate/action_model': 4.1662462591968905e-06, 'learning_rate/action_query_proj': 4.1662462591968905e-06, 'learning_rate/base': 1.0415615647992226e-06, 'epoch': 0.84}) +[2026-06-30 10:49:59,410][starVLA.training.train_starvla][INFO] - Step 1948, Loss: {'train/loss': 0.01445755548775196, 'train/grad_norm_pre_clip': 2.697673797607422, 'timing/data': 0.00028771301731467247, 'timing/model': 1.2292805220931768, 'learning_rate/action_model': 4.160035454366566e-06, 'learning_rate/action_query_proj': 4.160035454366566e-06, 'learning_rate/base': 1.0400088635916416e-06, 'epoch': 0.84}) +[2026-06-30 10:50:32,092][starVLA.training.train_starvla][INFO] - Step 1949, Loss: {'train/loss': 0.0005231369286775589, 'train/grad_norm_pre_clip': 4.407591819763184, 'timing/data': 0.00027347588911652565, 'timing/model': 1.2339213751256466, 'learning_rate/action_model': 4.153942689893123e-06, 'learning_rate/action_query_proj': 4.153942689893123e-06, 'learning_rate/base': 1.0384856724732807e-06, 'epoch': 0.84}) +[2026-06-30 10:51:04,776][starVLA.training.train_starvla][INFO] - Step 1950, Loss: {'train/loss': 5.1136394176864997e-05, 'train/grad_norm_pre_clip': 2.0747172832489014, 'timing/data': 0.00019554910250008106, 'timing/model': 1.2294154439587146, 'learning_rate/action_model': 4.147967980809858e-06, 'learning_rate/action_query_proj': 4.147967980809858e-06, 'learning_rate/base': 1.0369919952024644e-06, 'epoch': 0.84}) +[2026-06-30 10:51:37,466][starVLA.training.train_starvla][INFO] - Step 1951, Loss: {'train/loss': 0.09052270650863647, 'train/grad_norm_pre_clip': 3.945235013961792, 'timing/data': 0.00029944488778710365, 'timing/model': 1.228245756123215, 'learning_rate/action_model': 4.14211134185877e-06, 'learning_rate/action_query_proj': 4.14211134185877e-06, 'learning_rate/base': 1.0355278354646926e-06, 'epoch': 0.84}) +[2026-06-30 10:52:10,164][starVLA.training.train_starvla][INFO] - Step 1952, Loss: {'train/loss': 0.021875079721212387, 'train/grad_norm_pre_clip': 2.292595386505127, 'timing/data': 0.00025943899527192116, 'timing/model': 1.227688463870436, 'learning_rate/action_model': 4.136372787490531e-06, 'learning_rate/action_query_proj': 4.136372787490531e-06, 'learning_rate/base': 1.0340931968726327e-06, 'epoch': 0.84}) +[2026-06-30 10:52:42,887][starVLA.training.train_starvla][INFO] - Step 1953, Loss: {'train/loss': 0.0070992326363921165, 'train/grad_norm_pre_clip': 2.119805097579956, 'timing/data': 0.00033466098830103874, 'timing/model': 1.2310324299614877, 'learning_rate/action_model': 4.130752331864458e-06, 'learning_rate/action_query_proj': 4.130752331864458e-06, 'learning_rate/base': 1.0326880829661146e-06, 'epoch': 0.84}) +[2026-06-30 10:53:15,566][starVLA.training.train_starvla][INFO] - Step 1954, Loss: {'train/loss': 0.004141802433878183, 'train/grad_norm_pre_clip': 2.5746021270751953, 'timing/data': 0.00019866391085088253, 'timing/model': 1.2300576900597662, 'learning_rate/action_model': 4.125249988848462e-06, 'learning_rate/action_query_proj': 4.125249988848462e-06, 'learning_rate/base': 1.0313124972121156e-06, 'epoch': 0.84}) +[2026-06-30 10:53:48,235][starVLA.training.train_starvla][INFO] - Step 1955, Loss: {'train/loss': 7.986973287188448e-06, 'train/grad_norm_pre_clip': 2.492135763168335, 'timing/data': 0.00019649090245366096, 'timing/model': 1.23152811313048, 'learning_rate/action_model': 4.119865772019037e-06, 'learning_rate/action_query_proj': 4.119865772019037e-06, 'learning_rate/base': 1.0299664430047593e-06, 'epoch': 0.84}) +[2026-06-30 10:54:20,951][starVLA.training.train_starvla][INFO] - Step 1956, Loss: {'train/loss': 0.0002695386065170169, 'train/grad_norm_pre_clip': 2.6642725467681885, 'timing/data': 0.0003370661288499832, 'timing/model': 1.2254908438771963, 'learning_rate/action_model': 4.114599694661188e-06, 'learning_rate/action_query_proj': 4.114599694661188e-06, 'learning_rate/base': 1.028649923665297e-06, 'epoch': 0.84}) +[2026-06-30 10:54:53,613][starVLA.training.train_starvla][INFO] - Step 1957, Loss: {'train/loss': 0.40849167108535767, 'train/grad_norm_pre_clip': 2.3087050914764404, 'timing/data': 0.0001929940190166235, 'timing/model': 1.228271597996354, 'learning_rate/action_model': 4.109451769768448e-06, 'learning_rate/action_query_proj': 4.109451769768448e-06, 'learning_rate/base': 1.027362942442112e-06, 'epoch': 0.84}) +[2026-06-30 10:55:26,292][starVLA.training.train_starvla][INFO] - Step 1958, Loss: {'train/loss': 3.8234480598475784e-05, 'train/grad_norm_pre_clip': 2.0478782653808594, 'timing/data': 0.00019692094065248966, 'timing/model': 1.231812787009403, 'learning_rate/action_model': 4.104422010042808e-06, 'learning_rate/action_query_proj': 4.104422010042808e-06, 'learning_rate/base': 1.026105502510702e-06, 'epoch': 0.84}) +[2026-06-30 10:55:58,953][starVLA.training.train_starvla][INFO] - Step 1959, Loss: {'train/loss': 0.3734988272190094, 'train/grad_norm_pre_clip': 3.589116096496582, 'timing/data': 0.00019773305393755436, 'timing/model': 1.2316949740052223, 'learning_rate/action_model': 4.099510427894701e-06, 'learning_rate/action_query_proj': 4.099510427894701e-06, 'learning_rate/base': 1.0248776069736752e-06, 'epoch': 0.84}) +[2026-06-30 10:56:31,622][starVLA.training.train_starvla][INFO] - Step 1960, Loss: {'train/loss': 0.010423630475997925, 'train/grad_norm_pre_clip': 2.9497547149658203, 'timing/data': 0.00021099811419844627, 'timing/model': 1.2315450909081846, 'learning_rate/action_model': 4.094717035442964e-06, 'learning_rate/action_query_proj': 4.094717035442964e-06, 'learning_rate/base': 1.023679258860741e-06, 'epoch': 0.84}) +[2026-06-30 10:57:04,300][starVLA.training.train_starvla][INFO] - Step 1961, Loss: {'train/loss': 0.0005344520322978497, 'train/grad_norm_pre_clip': 2.1675028800964355, 'timing/data': 0.0002533779479563236, 'timing/model': 1.2287935900967568, 'learning_rate/action_model': 4.090041844514821e-06, 'learning_rate/action_query_proj': 4.090041844514821e-06, 'learning_rate/base': 1.0225104611287053e-06, 'epoch': 0.84}) +[2026-06-30 10:57:36,975][starVLA.training.train_starvla][INFO] - Step 1962, Loss: {'train/loss': 0.13764475286006927, 'train/grad_norm_pre_clip': 1.7676323652267456, 'timing/data': 0.00035070208832621574, 'timing/model': 1.2258072670083493, 'learning_rate/action_model': 4.085484866645833e-06, 'learning_rate/action_query_proj': 4.085484866645833e-06, 'learning_rate/base': 1.0213712166614583e-06, 'epoch': 0.84}) +[2026-06-30 10:58:09,655][starVLA.training.train_starvla][INFO] - Step 1963, Loss: {'train/loss': 0.002027446636930108, 'train/grad_norm_pre_clip': 4.807971954345703, 'timing/data': 0.00019463710486888885, 'timing/model': 1.2270222480874509, 'learning_rate/action_model': 4.081046113079901e-06, 'learning_rate/action_query_proj': 4.081046113079901e-06, 'learning_rate/base': 1.0202615282699752e-06, 'epoch': 0.84}) +[2026-06-30 10:58:42,269][starVLA.training.train_starvla][INFO] - Step 1964, Loss: {'train/loss': 0.00016677317034918815, 'train/grad_norm_pre_clip': 1.9391387701034546, 'timing/data': 0.00023537385277450085, 'timing/model': 1.228781124111265, 'learning_rate/action_model': 4.076725594769199e-06, 'learning_rate/action_query_proj': 4.076725594769199e-06, 'learning_rate/base': 1.0191813986922997e-06, 'epoch': 0.85}) +[2026-06-30 10:59:14,938][starVLA.training.train_starvla][INFO] - Step 1965, Loss: {'train/loss': 0.0001290872023673728, 'train/grad_norm_pre_clip': 2.021230697631836, 'timing/data': 0.0002317270264029503, 'timing/model': 1.2321126791648567, 'learning_rate/action_model': 4.0725233223741794e-06, 'learning_rate/action_query_proj': 4.0725233223741794e-06, 'learning_rate/base': 1.0181308305935449e-06, 'epoch': 0.85}) +[2026-06-30 10:59:47,594][starVLA.training.train_starvla][INFO] - Step 1966, Loss: {'train/loss': 1.9211161136627197, 'train/grad_norm_pre_clip': 5.472785949707031, 'timing/data': 0.00020317314192652702, 'timing/model': 1.229447390185669, 'learning_rate/action_model': 4.068439306263535e-06, 'learning_rate/action_query_proj': 4.068439306263535e-06, 'learning_rate/base': 1.0171098265658837e-06, 'epoch': 0.85}) +[2026-06-30 11:00:20,261][starVLA.training.train_starvla][INFO] - Step 1967, Loss: {'train/loss': 0.006953668314963579, 'train/grad_norm_pre_clip': 1.502632737159729, 'timing/data': 0.00032083503901958466, 'timing/model': 1.2288147069048136, 'learning_rate/action_model': 4.06447355651416e-06, 'learning_rate/action_query_proj': 4.06447355651416e-06, 'learning_rate/base': 1.01611838912854e-06, 'epoch': 0.85}) +[2026-06-30 11:00:52,903][starVLA.training.train_starvla][INFO] - Step 1968, Loss: {'train/loss': 0.27001088857650757, 'train/grad_norm_pre_clip': 1.6431350708007812, 'timing/data': 0.00019681104458868504, 'timing/model': 1.231208330951631, 'learning_rate/action_model': 4.060626082911161e-06, 'learning_rate/action_query_proj': 4.060626082911161e-06, 'learning_rate/base': 1.0151565207277903e-06, 'epoch': 0.85}) +[2026-06-30 11:01:25,558][starVLA.training.train_starvla][INFO] - Step 1969, Loss: {'train/loss': 0.00010384085908299312, 'train/grad_norm_pre_clip': 2.2403223514556885, 'timing/data': 0.00020221085287630558, 'timing/model': 1.2303906390443444, 'learning_rate/action_model': 4.056896894947787e-06, 'learning_rate/action_query_proj': 4.056896894947787e-06, 'learning_rate/base': 1.0142242237369467e-06, 'epoch': 0.85}) +[2026-06-30 11:01:58,218][starVLA.training.train_starvla][INFO] - Step 1970, Loss: {'train/loss': 0.02567942999303341, 'train/grad_norm_pre_clip': 2.1866137981414795, 'timing/data': 0.00019320379942655563, 'timing/model': 1.23013074696064, 'learning_rate/action_model': 4.053286001825445e-06, 'learning_rate/action_query_proj': 4.053286001825445e-06, 'learning_rate/base': 1.0133215004563612e-06, 'epoch': 0.85}) +[2026-06-30 11:02:30,869][starVLA.training.train_starvla][INFO] - Step 1971, Loss: {'train/loss': 0.014673730358481407, 'train/grad_norm_pre_clip': 1.7495760917663574, 'timing/data': 0.0002152861561626196, 'timing/model': 1.2324112998321652, 'learning_rate/action_model': 4.049793412453642e-06, 'learning_rate/action_query_proj': 4.049793412453642e-06, 'learning_rate/base': 1.0124483531134105e-06, 'epoch': 0.85}) +[2026-06-30 11:03:03,659][starVLA.training.train_starvla][INFO] - Step 1972, Loss: {'train/loss': 0.5709679126739502, 'train/grad_norm_pre_clip': 2.636460065841675, 'timing/data': 0.0002237320877611637, 'timing/model': 1.2279358070809394, 'learning_rate/action_model': 4.046419135450014e-06, 'learning_rate/action_query_proj': 4.046419135450014e-06, 'learning_rate/base': 1.0116047838625034e-06, 'epoch': 0.85}) +[2026-06-30 11:03:36,339][starVLA.training.train_starvla][INFO] - Step 1973, Loss: {'train/loss': 0.3453753888607025, 'train/grad_norm_pre_clip': 4.745983600616455, 'timing/data': 0.0003085620701313019, 'timing/model': 1.2299740621820092, 'learning_rate/action_model': 4.043163179140242e-06, 'learning_rate/action_query_proj': 4.043163179140242e-06, 'learning_rate/base': 1.0107907947850604e-06, 'epoch': 0.85}) +[2026-06-30 11:04:08,984][starVLA.training.train_starvla][INFO] - Step 1974, Loss: {'train/loss': 0.003969179932028055, 'train/grad_norm_pre_clip': 1.7444732189178467, 'timing/data': 0.00021193991415202618, 'timing/model': 1.2277795220725238, 'learning_rate/action_model': 4.040025551558075e-06, 'learning_rate/action_query_proj': 4.040025551558075e-06, 'learning_rate/base': 1.0100063878895187e-06, 'epoch': 0.85}) +[2026-06-30 11:04:41,625][starVLA.training.train_starvla][INFO] - Step 1975, Loss: {'train/loss': 0.008742894046008587, 'train/grad_norm_pre_clip': 4.605983734130859, 'timing/data': 0.00036406703293323517, 'timing/model': 1.2300622570328414, 'learning_rate/action_model': 4.0370062604453e-06, 'learning_rate/action_query_proj': 4.0370062604453e-06, 'learning_rate/base': 1.009251565111325e-06, 'epoch': 0.85}) +[2026-06-30 11:05:14,268][starVLA.training.train_starvla][INFO] - Step 1976, Loss: {'train/loss': 0.006637281272560358, 'train/grad_norm_pre_clip': 3.51798939704895, 'timing/data': 0.0002525760792195797, 'timing/model': 1.2272502130363137, 'learning_rate/action_model': 4.034105313251717e-06, 'learning_rate/action_query_proj': 4.034105313251717e-06, 'learning_rate/base': 1.0085263283129293e-06, 'epoch': 0.85}) +[2026-06-30 11:05:46,922][starVLA.training.train_starvla][INFO] - Step 1977, Loss: {'train/loss': 0.1989804208278656, 'train/grad_norm_pre_clip': 2.914720296859741, 'timing/data': 0.00019677099771797657, 'timing/model': 1.2279383200220764, 'learning_rate/action_model': 4.031322717135122e-06, 'learning_rate/action_query_proj': 4.031322717135122e-06, 'learning_rate/base': 1.0078306792837805e-06, 'epoch': 0.85}) +[2026-06-30 11:06:19,555][starVLA.training.train_starvla][INFO] - Step 1978, Loss: {'train/loss': 1.4662492503703106e-05, 'train/grad_norm_pre_clip': 1.9838966131210327, 'timing/data': 0.0002627461217343807, 'timing/model': 1.2300261280033737, 'learning_rate/action_model': 4.0286584789613e-06, 'learning_rate/action_query_proj': 4.0286584789613e-06, 'learning_rate/base': 1.007164619740325e-06, 'epoch': 0.85}) +[2026-06-30 11:06:52,206][starVLA.training.train_starvla][INFO] - Step 1979, Loss: {'train/loss': 0.032358694821596146, 'train/grad_norm_pre_clip': 1.771693468093872, 'timing/data': 0.00021317205391824245, 'timing/model': 1.2317122279200703, 'learning_rate/action_model': 4.026112605303993e-06, 'learning_rate/action_query_proj': 4.026112605303993e-06, 'learning_rate/base': 1.0065281513259982e-06, 'epoch': 0.85}) +[2026-06-30 11:07:24,857][starVLA.training.train_starvla][INFO] - Step 1980, Loss: {'train/loss': 0.00073088682256639, 'train/grad_norm_pre_clip': 1.3348716497421265, 'timing/data': 0.00020068790763616562, 'timing/model': 1.2293467270210385, 'learning_rate/action_model': 4.023685102444883e-06, 'learning_rate/action_query_proj': 4.023685102444883e-06, 'learning_rate/base': 1.0059212756112208e-06, 'epoch': 0.85}) +[2026-06-30 11:07:57,520][starVLA.training.train_starvla][INFO] - Step 1981, Loss: {'train/loss': 0.04630206897854805, 'train/grad_norm_pre_clip': 1.9331446886062622, 'timing/data': 0.00019607995636761189, 'timing/model': 1.2319642440415919, 'learning_rate/action_model': 4.021375976373601e-06, 'learning_rate/action_query_proj': 4.021375976373601e-06, 'learning_rate/base': 1.0053439940934003e-06, 'epoch': 0.85}) +[2026-06-30 11:08:30,199][starVLA.training.train_starvla][INFO] - Step 1982, Loss: {'train/loss': 0.4327358305454254, 'train/grad_norm_pre_clip': 2.135822057723999, 'timing/data': 0.00026412797160446644, 'timing/model': 1.2265669831540436, 'learning_rate/action_model': 4.019185232787686e-06, 'learning_rate/action_query_proj': 4.019185232787686e-06, 'learning_rate/base': 1.0047963081969214e-06, 'epoch': 0.85}) +[2026-06-30 11:09:02,844][starVLA.training.train_starvla][INFO] - Step 1983, Loss: {'train/loss': 0.0027031656354665756, 'train/grad_norm_pre_clip': 2.876908779144287, 'timing/data': 0.0002499909605830908, 'timing/model': 1.2294418760575354, 'learning_rate/action_model': 4.017112877092573e-06, 'learning_rate/action_query_proj': 4.017112877092573e-06, 'learning_rate/base': 1.0042782192731433e-06, 'epoch': 0.85}) +[2026-06-30 11:09:35,458][starVLA.training.train_starvla][INFO] - Step 1984, Loss: {'train/loss': 0.126898854970932, 'train/grad_norm_pre_clip': 1.5003836154937744, 'timing/data': 0.00020653009414672852, 'timing/model': 1.2330718620214611, 'learning_rate/action_model': 4.015158914401603e-06, 'learning_rate/action_query_proj': 4.015158914401603e-06, 'learning_rate/base': 1.0037897286004007e-06, 'epoch': 0.85}) +[2026-06-30 11:10:08,112][starVLA.training.train_starvla][INFO] - Step 1985, Loss: {'train/loss': 9.68569474935066e-06, 'train/grad_norm_pre_clip': 1.600045919418335, 'timing/data': 0.00020756199955940247, 'timing/model': 1.2311631108168513, 'learning_rate/action_model': 4.013323349535973e-06, 'learning_rate/action_query_proj': 4.013323349535973e-06, 'learning_rate/base': 1.0033308373839932e-06, 'epoch': 0.85}) +[2026-06-30 11:10:40,810][starVLA.training.train_starvla][INFO] - Step 1986, Loss: {'train/loss': 0.16231301426887512, 'train/grad_norm_pre_clip': 1.043761134147644, 'timing/data': 0.0003665308468043804, 'timing/model': 1.2298539918847382, 'learning_rate/action_model': 4.011606187024766e-06, 'learning_rate/action_query_proj': 4.011606187024766e-06, 'learning_rate/base': 1.0029015467561915e-06, 'epoch': 0.85}) +[2026-06-30 11:11:13,476][starVLA.training.train_starvla][INFO] - Step 1987, Loss: {'train/loss': 1.2794907093048096, 'train/grad_norm_pre_clip': 3.3406920433044434, 'timing/data': 0.00024949992075562477, 'timing/model': 1.2298471690155566, 'learning_rate/action_model': 4.010007431104906e-06, 'learning_rate/action_query_proj': 4.010007431104906e-06, 'learning_rate/base': 1.0025018577762265e-06, 'epoch': 0.85}) +[2026-06-30 11:11:46,186][starVLA.training.train_starvla][INFO] - Step 1988, Loss: {'train/loss': 0.00034154753666371107, 'train/grad_norm_pre_clip': 2.2777938842773438, 'timing/data': 0.00025066290982067585, 'timing/model': 1.229212170932442, 'learning_rate/action_model': 4.0085270857211695e-06, 'learning_rate/action_query_proj': 4.0085270857211695e-06, 'learning_rate/base': 1.0021317714302924e-06, 'epoch': 0.86}) +[2026-06-30 11:12:18,847][starVLA.training.train_starvla][INFO] - Step 1989, Loss: {'train/loss': 1.1235370948270429e-05, 'train/grad_norm_pre_clip': 2.300076961517334, 'timing/data': 0.00020659994333982468, 'timing/model': 1.2375621050596237, 'learning_rate/action_model': 4.007165154526154e-06, 'learning_rate/action_query_proj': 4.007165154526154e-06, 'learning_rate/base': 1.0017912886315385e-06, 'epoch': 0.86}) +[2026-06-30 11:12:51,557][starVLA.training.train_starvla][INFO] - Step 1990, Loss: {'train/loss': 0.38386863470077515, 'train/grad_norm_pre_clip': 3.3246419429779053, 'timing/data': 0.0002723031211644411, 'timing/model': 1.2311591021716595, 'learning_rate/action_model': 4.0059216408802915e-06, 'learning_rate/action_query_proj': 4.0059216408802915e-06, 'learning_rate/base': 1.0014804102200729e-06, 'epoch': 0.86}) +[2026-06-30 11:13:24,236][starVLA.training.train_starvla][INFO] - Step 1991, Loss: {'train/loss': 5.960438102192711e-06, 'train/grad_norm_pre_clip': 1.9013558626174927, 'timing/data': 0.0001992459874600172, 'timing/model': 1.2281297841109335, 'learning_rate/action_model': 4.004796547851832e-06, 'learning_rate/action_query_proj': 4.004796547851832e-06, 'learning_rate/base': 1.001199136962958e-06, 'epoch': 0.86}) +[2026-06-30 11:13:56,897][starVLA.training.train_starvla][INFO] - Step 1992, Loss: {'train/loss': 0.06948280334472656, 'train/grad_norm_pre_clip': 2.119929790496826, 'timing/data': 0.0002718220930546522, 'timing/model': 1.2302588960155845, 'learning_rate/action_model': 4.0037898782168265e-06, 'learning_rate/action_query_proj': 4.0037898782168265e-06, 'learning_rate/base': 1.0009474695542066e-06, 'epoch': 0.86}) +[2026-06-30 11:14:29,626][starVLA.training.train_starvla][INFO] - Step 1993, Loss: {'train/loss': 7.003514383541187e-06, 'train/grad_norm_pre_clip': 6.540798187255859, 'timing/data': 0.0001998769585043192, 'timing/model': 1.2279409898910671, 'learning_rate/action_model': 4.002901634459133e-06, 'learning_rate/action_query_proj': 4.002901634459133e-06, 'learning_rate/base': 1.0007254086147833e-06, 'epoch': 0.86}) +[2026-06-30 11:15:02,290][starVLA.training.train_starvla][INFO] - Step 1994, Loss: {'train/loss': 0.10104950517416, 'train/grad_norm_pre_clip': 1.7812577486038208, 'timing/data': 0.0002753890585154295, 'timing/model': 1.233324731932953, 'learning_rate/action_model': 4.002131818770408e-06, 'learning_rate/action_query_proj': 4.002131818770408e-06, 'learning_rate/base': 1.000532954692602e-06, 'epoch': 0.86}) +[2026-06-30 11:15:34,916][starVLA.training.train_starvla][INFO] - Step 1995, Loss: {'train/loss': 0.009759511798620224, 'train/grad_norm_pre_clip': 2.186669111251831, 'timing/data': 0.00019647087901830673, 'timing/model': 1.2327291457913816, 'learning_rate/action_model': 4.0014804330500954e-06, 'learning_rate/action_query_proj': 4.0014804330500954e-06, 'learning_rate/base': 1.0003701082625239e-06, 'epoch': 0.86}) +[2026-06-30 11:16:07,610][starVLA.training.train_starvla][INFO] - Step 1996, Loss: {'train/loss': 0.04708588868379593, 'train/grad_norm_pre_clip': 2.2413275241851807, 'timing/data': 0.00024981098249554634, 'timing/model': 1.2302646541502327, 'learning_rate/action_model': 4.000947478905418e-06, 'learning_rate/action_query_proj': 4.000947478905418e-06, 'learning_rate/base': 1.0002368697263545e-06, 'epoch': 0.86}) +[2026-06-30 11:16:40,266][starVLA.training.train_starvla][INFO] - Step 1997, Loss: {'train/loss': 0.2814132273197174, 'train/grad_norm_pre_clip': 3.6154251098632812, 'timing/data': 0.0002980118151754141, 'timing/model': 1.2301610249560326, 'learning_rate/action_model': 4.000532957651392e-06, 'learning_rate/action_query_proj': 4.000532957651392e-06, 'learning_rate/base': 1.000133239412848e-06, 'epoch': 0.86}) +[2026-06-30 11:17:12,864][starVLA.training.train_starvla][INFO] - Step 1998, Loss: {'train/loss': 0.2972881495952606, 'train/grad_norm_pre_clip': 1.762986183166504, 'timing/data': 0.00025204592384397984, 'timing/model': 1.2278200171422213, 'learning_rate/action_model': 4.0002368703108056e-06, 'learning_rate/action_query_proj': 4.0002368703108056e-06, 'learning_rate/base': 1.0000592175777014e-06, 'epoch': 0.86}) +[2026-06-30 11:17:45,470][starVLA.training.train_starvla][INFO] - Step 1999, Loss: {'train/loss': 0.0006116971489973366, 'train/grad_norm_pre_clip': 3.7971391677856445, 'timing/data': 0.00019488786347210407, 'timing/model': 1.2298179029021412, 'learning_rate/action_model': 4.000059217614231e-06, 'learning_rate/action_query_proj': 4.000059217614231e-06, 'learning_rate/base': 1.0000148044035577e-06, 'epoch': 0.86}) +[2026-06-30 11:18:18,125][starVLA.training.train_starvla][INFO] - Starting held-out action loss eval at step 2000 +[2026-06-30 11:18:55,303][starVLA.training.train_starvla][INFO] - Finished held-out action loss eval at step 2000 +[2026-06-30 11:18:55,303][starVLA.training.train_starvla][INFO] - Starting held-out action classification eval at step 2000 +[2026-06-30 11:18:55,303][starVLA.training.train_starvla][INFO] - Finished held-out action classification eval at step 2000 +[2026-06-30 12:07:29,205][starVLA.training.rl_games.checkpoint_sync][INFO] - HF checkpoint sync disabled; skipping eval result upload for step 2000 +[2026-06-30 12:07:29,205][starVLA.training.train_starvla][INFO] - Step 2000, Loss: {'train/loss': 0.10021309554576874, 'train/grad_norm_pre_clip': 2.1859207153320312, 'eval/action_loss/seconds': 37.175399794010445, 'eval/action_loss/samples': 200.0, 'eval/loss': 0.02366216853260994, 'eval/latency_1/loss': 0.02366216853260994, 'eval/flappy/loss': 0.02366216853260994, 'eval/flappy/latency_1/loss': 0.02366216853260994, 'rl_games_eval/mid_train/total_episodes': 5.0, 'rl_games_eval/mid_train/mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/mean_length': 3600.0, 'rl_games_eval/mid_train/std_reward': 0.0, 'rl_games_eval/mid_train/std_length': 0.0, 'rl_games_eval/mid_train/macro_mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/macro_mean_length': 3600.0, 'rl_games_eval/mid_train/task_count': 1.0, 'rl_games_eval/mid_train/flappy__latency_1/mean_reward': 444.60000000002566, 'rl_games_eval/mid_train/flappy__latency_1/mean_length': 3600.0, 'rl_games_eval/mid_train/flappy__latency_1/std_reward': 0.0, 'rl_games_eval/mid_train/flappy__latency_1/std_length': 0.0, 'rl_games_eval/mid_train/flappy__latency_1/num_episodes': 5.0, 'checkpoint/current_eval_score': 444.60000000002566, 'checkpoint/best_score': -inf, 'checkpoint/best_step': 0.0, 'checkpoint/is_new_best': 0.0, 'timing/data': 0.0003980710171163082, 'timing/model': 1.2341117900796235, 'learning_rate/action_model': 4e-06, 'learning_rate/action_query_proj': 4e-06, 'learning_rate/base': 1e-06, 'epoch': 0.86}) +[2026-06-30 12:07:37,163][starVLA.training.rl_games.checkpoint_sync][INFO] - HF checkpoint sync disabled; skipping upload for step 2000 +[2026-06-30 12:07:37,163][starVLA.training.train_starvla][INFO] - 📊 Saving accessed configuration... +[2026-06-30 12:07:37,179][starVLA.training.train_starvla][INFO] - ✅ Configuration files saved +[2026-06-30 12:07:37,180][starVLA.training.train_starvla][INFO] - Training complete. Final training state is saved under /workspace/starVLA/results/Checkpoints/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/checkpoints +[2026-06-30 16:04:00,102][starVLA.training.rl_games.checkpoint_sync][INFO] - HF checkpoint sync disabled; skipping eval result upload for step 2000 +[2026-06-30 16:04:02,280][starVLA.training.train_starvla][INFO] - ... and that's all, folks! diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/summary.jsonl b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/summary.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..b1cc9ca82e5144b44a69491d6f0a069a7dc61811 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/duplicate_aliases/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/summary.jsonl @@ -0,0 +1 @@ +{"steps": 2000} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/fixed_conversion_probe.log b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/fixed_conversion_probe.log new file mode 100644 index 0000000000000000000000000000000000000000..3d0662f28115bed39351729374d7c1587bf89947 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/fixed_conversion_probe.log @@ -0,0 +1,9 @@ +[setup] rebuilding converted dataset because: +[setup] - dataset.setup_force or dataset.force_download is true +[setup] - train dataset is not ready: /tmp/experiment-repo-cleanup-20261007/conversion_probe/converted/flappy_train__bridge +[setup] - eval dataset is not ready: /tmp/experiment-repo-cleanup-20261007/conversion_probe/converted/flappy_train__bridge__val + Indexing Flappy train local parquet rows: 0it [00:00, ?it/s] Indexing Flappy train local parquet rows: 6it [00:00, 871.66it/s] + Writing Flappy train local parquet rows: 0it [00:00, ?it/s] Writing Flappy train local parquet rows: 6it [00:00, 975.84it/s] + Indexing Flappy validation local parquet rows: 0it [00:00, ?it/s] Indexing Flappy validation local parquet rows: 3it [00:00, 2777.07it/s] + Writing Flappy validation local parquet rows: 0it [00:00, ?it/s] Writing Flappy validation local parquet rows: 3it [00:00, 3028.38it/s] +{"fixed_revision_preserved": true, "train_and_val_converted": true, "actual_parquet_read": true, "snapshot_prefix": "training/flappy_200ep/flappy_fix_latency_2_200ep", "snapshot_calls": 1} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/native_loader_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/native_loader_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..ac0ad7f2c2d21dc29a697f5f14ac60998f6cc1f0 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/native_loader_verification.json @@ -0,0 +1,1866 @@ +[ + { + "source_repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_2", + "source_revision": "0a60d6c8a400b48e4d9737b36ef3a44814d52d2a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_deadly_corridor_fix_latency_2/checkpoints/best_model.safetensors", + "identity": "630d55bc258d986e0de4b04d9223f418855bb94c2d310c13ae808912553d6c7f", + "runpath": "historical/openvla_deadly_corridor_fix_latency_2", + "configpath": "historical/openvla_deadly_corridor_fix_latency_2/config.yaml", + "statpath": "historical/openvla_deadly_corridor_fix_latency_2/dataset_statistics.json", + "run_id": "openvla_deadly_corridor_fix_latency_2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0_full", + "source_revision": "9281fd061894c14d3641ee9724e108cd33baf494", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_deadly_corridor_fix_latency_0_full/checkpoints/best_model.safetensors", + "identity": "5de650dbc0c50201489cb391a18f95b76bc1b9fb5a3df03ee965050b90284e84", + "runpath": "historical/openvla_deadly_corridor_fix_latency_0_full", + "configpath": "historical/openvla_deadly_corridor_fix_latency_0_full/config.yaml", + "statpath": "historical/openvla_deadly_corridor_fix_latency_0_full/dataset_statistics.json", + "run_id": "openvla_deadly_corridor_fix_latency_0", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "914436f584ae969603a7e9b617a400f266e7fb9f", + "917d564a90c8209144f9b50204c47192f27e1ac1", + "9e1cf489a76ed2c07abc29399d637d5d73417ccb", + "d9dfe4578bc891fe65984d0daaad1e9783ac51de" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0_small", + "source_revision": "05586f38706727534a0f388ae0ece46541776d65", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_0_small/checkpoints/best_model.safetensors", + "identity": "9117254092707a50375a50e48626afdfbc3e7ac1ea51320a09ec901cd53f0e37", + "runpath": "historical/openvla_flappy_fix_latency_0_small", + "configpath": "historical/openvla_flappy_fix_latency_0_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_0_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_0_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_6", + "source_revision": "7a2e2f81dda9fc9de068fa15b1eac442ed6fb51f", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_6/checkpoints/steps_5000_pytorch_model.pt", + "identity": "26943b407d9280f89a461ca674615037f1e2ddda896fe00b88015ff5ce22a9f0", + "runpath": "historical/openvla_flappy_fix_latency_6", + "configpath": "historical/openvla_flappy_fix_latency_6/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_6/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_6", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1e638dd5cb934c2e322f2f3879174c7d626e4410" + ], + "loader_revision": "1e638dd5cb934c2e322f2f3879174c7d626e4410", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1e638dd5cb934c2e322f2f3879174c7d626e4410", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_l2_full", + "source_revision": "2e8e58c6af1ee6f8a3701388b2552a4fa65a041c", + "repo": "latency-sensitive-bench/vla-finetuning-scope-models", + "revision": "c1275f13f74269e997f8f369d164931279f9014a", + "path": "historical/openvla_flappy_l2_full/final_model/pytorch_model.pt", + "identity": "f6be353e457ac9cf6dbda6cd2e93692d797016100b69ecea4bcab0ececb9a076", + "runpath": "historical/openvla_flappy_l2_full", + "configpath": "historical/openvla_flappy_l2_full/config.yaml", + "statpath": "historical/openvla_flappy_l2_full/dataset_statistics.json", + "run_id": "openvla_flappy_l2_full", + "kind": "full", + "native_ready": true, + "base_model": "action", + "framework": "QwenOFT", + "wandb_entity": "saberrr-zju", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "538fd523fe4fa4e7abce3201611919314bc6b664" + ], + "loader_revision": "538fd523fe4fa4e7abce3201611919314bc6b664", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/final_model/pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/538fd523fe4fa4e7abce3201611919314bc6b664", + "backbone": "/tmp/experiment-repo-validation-20261007/action-base-0/models--StarVLA--Qwen3-VL-4B-Instruct-Action/snapshots/c41500cf1d287cd79e9f6602f419937b05048bd8", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "demon_attack_fix_latency_0_200ep/checkpoints/steps_7000_model.safetensors", + "identity": "25a15d0460a63a7ca036e9300d39a4f57b8f9e656c6ff42941348b6a570bf767", + "runpath": "demon_attack_fix_latency_0_200ep", + "configpath": "demon_attack_fix_latency_0_200ep/config.yaml", + "statpath": "demon_attack_fix_latency_0_200ep/dataset_statistics.json", + "run_id": "demon_attack_fix_latency_0_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1a93a1ae6772e74087a836b7b2246d340b948bd2" + ], + "loader_revision": "1a93a1ae6772e74087a836b7b2246d340b948bd2", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_7000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1a93a1ae6772e74087a836b7b2246d340b948bd2", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "demon_attack_fix_latency_4_200ep/checkpoints/steps_7000_model.safetensors", + "identity": "62e9b2a2b38f2b52cedc248c6e135b2593b212d1965fc3efae42e8130b57b373", + "runpath": "demon_attack_fix_latency_4_200ep", + "configpath": "demon_attack_fix_latency_4_200ep/config.yaml", + "statpath": "demon_attack_fix_latency_4_200ep/dataset_statistics.json", + "run_id": "demon_attack_fix_latency_4_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + ], + "loader_revision": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_7000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "demon_attack_fix_latency_8_200ep/checkpoints/steps_7000_model.safetensors", + "identity": "0d52d791412c41f17f607087c988d99267ef09ac097022b3fc3d97ebbb5b0b00", + "runpath": "demon_attack_fix_latency_8_200ep", + "configpath": "demon_attack_fix_latency_8_200ep/config.yaml", + "statpath": "demon_attack_fix_latency_8_200ep/dataset_statistics.json", + "run_id": "demon_attack_fix_latency_8_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + ], + "loader_revision": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_7000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "source_revision": "5dfd90884fc4e3817bc51f574d6fc9cea2d7948b", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3/checkpoints/steps_4000_model.safetensors", + "identity": "553d47011aa16cef4609610ea44342630fb4b0b7aabc65da5fe58d133915f40a", + "runpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "configpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3/config.yaml", + "statpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3/dataset_statistics.json", + "run_id": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_4000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_3_small", + "source_revision": "86b4d314c377420fa8eb9ac7aa268695c2ab4f40", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_3_small/checkpoints/steps_1500_model.safetensors", + "identity": "8b61cc767d8a885e8bf0e3fecc383c8e3233beb3f3ad2544e6f7c8c4265ae614", + "runpath": "historical/openvla_flappy_fix_latency_3_small", + "configpath": "historical/openvla_flappy_fix_latency_3_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_3_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_3_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_1500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small", + "source_revision": "7842a04e6ba8ed4dc56ccdae235735a807a77b5d", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_1_small/checkpoints/steps_1500_model.safetensors", + "identity": "0ecf34ec797bcb608fabd5fee1b994bf8480dfd6d75eb302b9a5380cf9d69151", + "runpath": "historical/openvla_flappy_fix_latency_1_small", + "configpath": "historical/openvla_flappy_fix_latency_1_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_1_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_1_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_1500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small_tmp", + "source_revision": "8fba8c350f6f10d70119807a546218cfa82e9150", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_1_small_tmp/checkpoints/steps_1500_model.safetensors", + "identity": "76504a6629f8a781c5cfdbeac1795bdc6e221ba0d1931eabaf12d5bf5d56ccc1", + "runpath": "historical/openvla_flappy_fix_latency_1_small_tmp", + "configpath": "historical/openvla_flappy_fix_latency_1_small_tmp/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_1_small_tmp/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_1", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "0bc71a07a07274d35bc9aa60f8d8d14306d45c24", + "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_1500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3", + "source_revision": "f8d64aff21628adbdd0d1aaddd1bc7067bc2a23c", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3/checkpoints/steps_4000_model.safetensors", + "identity": "834ce01a8a1ec7ec4da4b6ea2765d7dbc249af3f80e59c571744dcee4242a13e", + "runpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3", + "configpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3/config.yaml", + "statpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3/dataset_statistics.json", + "run_id": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_4000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_small", + "source_revision": "371074f07545d1d4e5533afc4d457a63d03bf9da", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_4_small/checkpoints/steps_1500_model.safetensors", + "identity": "e3af4a77ec1ff9bb6d5cc58129a8e8b5b060d80d978228564561e188791d8029", + "runpath": "historical/openvla_flappy_fix_latency_4_small", + "configpath": "historical/openvla_flappy_fix_latency_4_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_4_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_4_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_1500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2_small", + "source_revision": "87c38afed1a227e92aac9efeb0921f57d556df8a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_2_small/checkpoints/steps_1500_model.safetensors", + "identity": "3aa56b5e881919571003b6155387d5ea5dfe50181e8f57b788e15e94dc643c73", + "runpath": "historical/openvla_flappy_fix_latency_2_small", + "configpath": "historical/openvla_flappy_fix_latency_2_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_2_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_2_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "d182480d774978c8e37efb1d3ea983d812acd4f6" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_1500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2", + "source_revision": "8135db54520527abe4a62f002323a307d1a47c2d", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_2/checkpoints/steps_5000_model.safetensors", + "identity": "0fc5f0a0cd58f4f9726d53639e193c48efd5afd512fb0d0f644d239eb4079668", + "runpath": "historical/openvla_flappy_fix_latency_2", + "configpath": "historical/openvla_flappy_fix_latency_2/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_2/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1e638dd5cb934c2e322f2f3879174c7d626e4410", + "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_2_small", + "source_revision": "bb91ae40a1bb9b9f6a93be031193de603da86ecd", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_demon_attack_fix_latency_2_small/checkpoints/steps_2000_model.safetensors", + "identity": "adb853d16890002501675633ddefc2126fd3f102a63cf6ece7a5e56d0c8c662c", + "runpath": "historical/openvla_demon_attack_fix_latency_2_small", + "configpath": "historical/openvla_demon_attack_fix_latency_2_small/config.yaml", + "statpath": "historical/openvla_demon_attack_fix_latency_2_small/dataset_statistics.json", + "run_id": "openvla_demon_attack_fix_latency_2_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_2000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "source_revision": "cd810024c06f961c95449785f19ab73ccfd20de1", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15/checkpoints/steps_5000_model.safetensors", + "identity": "2506bd51c7054d65cc980738979dd484de07b4af15da554185cbc28ede512de5", + "runpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "configpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15/config.yaml", + "statpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15/dataset_statistics.json", + "run_id": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "source_revision": "bd6274f20b7ff2f36c0b758058ef8ce2e996d7f6", + "repo": "latency-sensitive-bench/latency-information-models", + "revision": "b19c1eca8caefeaafcfbfc39f53c240c4e9a255c", + "path": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2/checkpoints/steps_5000_pytorch_model.pt", + "identity": "d275016d512fe0fe071d5c2c9f7362fba939c89cb357e461d8b59e96c31ba62f", + "runpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "configpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2/config.yaml", + "statpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2/dataset_statistics.json", + "run_id": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "talha1503", + "wandb_project": "latency-sensitive-bench", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1", + "source_revision": "004ec237af79c10de5c0d4376cf8a13542ac01f8", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1/checkpoints/steps_5000_model.safetensors", + "identity": "91931822871b9c3af55f58a7254963b512e478bda6d7514dd2a194ac27311cd5", + "runpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1", + "configpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1/config.yaml", + "statpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1/dataset_statistics.json", + "run_id": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2", + "source_revision": "73ea97c064d206c3a0dcf6ffdbf2ecfe97bc2c3c", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2/checkpoints/steps_5000_model.safetensors", + "identity": "91931822871b9c3af55f58a7254963b512e478bda6d7514dd2a194ac27311cd5", + "runpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2", + "configpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2/config.yaml", + "statpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2/dataset_statistics.json", + "run_id": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "flappy_curriculum_exclusive_40ep_per_latency/checkpoints/steps_5000_model.safetensors", + "identity": "4a6b8739fd13e8de24eaacf3f1e133ea6792e01f66cd919ed980d744c0a52e4e", + "runpath": "flappy_curriculum_exclusive_40ep_per_latency", + "configpath": "flappy_curriculum_exclusive_40ep_per_latency/config.yaml", + "statpath": "flappy_curriculum_exclusive_40ep_per_latency/dataset_statistics.json", + "run_id": "flappy_curriculum_exclusive_40ep_per_latency", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "ba36a6b93440808752b66ba5d25281b4c0ce5a8b", + "d1a786ff88f4a5aff08baa76b8c82ab4964a2804" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "flappy_fix_latency_1_200ep/checkpoints/steps_5000_model.safetensors", + "identity": "cff93451f827160bf275aa8bc0bcab83cbd71645ef0605b2e8869bd3691a3bb6", + "runpath": "flappy_fix_latency_1_200ep", + "configpath": "flappy_fix_latency_1_200ep/config.yaml", + "statpath": "flappy_fix_latency_1_200ep/dataset_statistics.json", + "run_id": "flappy_fix_latency_1_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "d05200d75af235edfb5db5bccd1f870625e288c1" + ], + "loader_revision": "d05200d75af235edfb5db5bccd1f870625e288c1", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/d05200d75af235edfb5db5bccd1f870625e288c1", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "flappy_fix_latency_3_200ep/checkpoints/steps_5000_model.safetensors", + "identity": "5da01d6eeb7a46861da90b0a8dbc364a33f92d542d25079bcc86280b10617418", + "runpath": "flappy_fix_latency_3_200ep", + "configpath": "flappy_fix_latency_3_200ep/config.yaml", + "statpath": "flappy_fix_latency_3_200ep/dataset_statistics.json", + "run_id": "flappy_fix_latency_3_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "ffe48f30f636bd6984f9d8c37e9b54c9f474c172" + ], + "loader_revision": "ffe48f30f636bd6984f9d8c37e9b54c9f474c172", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/ffe48f30f636bd6984f9d8c37e9b54c9f474c172", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0", + "source_revision": "06f64d1a03905ad0f04a447c57afb970c29a559e", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_0/checkpoints/steps_5000_pytorch_model.pt", + "identity": "a7581596728be2fb496d24bea5ec1c8586a5fcbf8ee99f1e9d5c90f11288f70b", + "runpath": "historical/openvla_flappy_fix_latency_0", + "configpath": "historical/openvla_flappy_fix_latency_0/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_0/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_0", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "your_wandb_entity", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1", + "source_revision": "bc733d64e32512a8fc76f9be022eb5b0d7c0e2cb", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1/checkpoints/steps_5000_model.safetensors", + "identity": "2506bd51c7054d65cc980738979dd484de07b4af15da554185cbc28ede512de5", + "runpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1", + "configpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1/config.yaml", + "statpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1/dataset_statistics.json", + "run_id": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0", + "source_revision": "9f8c6842151405a82a3e852d2a00e27885db1e93", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_deadly_corridor_fix_latency_0/checkpoints/steps_500_model.safetensors", + "identity": "b1478d01499d37121fee35d4f9f1dc090f6e87192797f50556388c94970c3fc1", + "runpath": "historical/openvla_deadly_corridor_fix_latency_0", + "configpath": "historical/openvla_deadly_corridor_fix_latency_0/config.yaml", + "statpath": "historical/openvla_deadly_corridor_fix_latency_0/dataset_statistics.json", + "run_id": "openvla_deadly_corridor_fix_latency_0", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "914436f584ae969603a7e9b617a400f266e7fb9f", + "917d564a90c8209144f9b50204c47192f27e1ac1", + "9e1cf489a76ed2c07abc29399d637d5d73417ccb", + "d9dfe4578bc891fe65984d0daaad1e9783ac51de" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_exp", + "source_revision": "213bab6c06b13c73fee875412d99e40ea4a4dd70", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_4_exp/checkpoints/steps_2500_model.safetensors", + "identity": "b4a6e3ade6d47371998034284d7a0ccbeb1f92b9713a657eb6041604cdbe9235", + "runpath": "historical/openvla_flappy_fix_latency_4_exp", + "configpath": "historical/openvla_flappy_fix_latency_4_exp/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_4_exp/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_4_exp", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "46e36dafd933fd767313c1c84f7ace66c2683be2" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "compatibility_contract": "Documented Qwen tied LM-head checkpoint loading", + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_2500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_2", + "source_revision": "0a60d6c8a400b48e4d9737b36ef3a44814d52d2a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_deadly_corridor_fix_latency_2/checkpoints/steps_500_model.safetensors", + "identity": "ad5f732f19bbfda68d6686e39e875f07957d08980bf6325dea7f85eccb4abc70", + "runpath": "historical/openvla_deadly_corridor_fix_latency_2", + "configpath": "historical/openvla_deadly_corridor_fix_latency_2/config.yaml", + "statpath": "historical/openvla_deadly_corridor_fix_latency_2/dataset_statistics.json", + "run_id": "openvla_deadly_corridor_fix_latency_2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0_full", + "source_revision": "9281fd061894c14d3641ee9724e108cd33baf494", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_deadly_corridor_fix_latency_0_full/checkpoints/steps_2500_model.safetensors", + "identity": "dfdb46c577b89bc4dcdef272e2b9e1d590ac0331c23632284bae3980d419bd5e", + "runpath": "historical/openvla_deadly_corridor_fix_latency_0_full", + "configpath": "historical/openvla_deadly_corridor_fix_latency_0_full/config.yaml", + "statpath": "historical/openvla_deadly_corridor_fix_latency_0_full/dataset_statistics.json", + "run_id": "openvla_deadly_corridor_fix_latency_0", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "914436f584ae969603a7e9b617a400f266e7fb9f", + "917d564a90c8209144f9b50204c47192f27e1ac1", + "9e1cf489a76ed2c07abc29399d637d5d73417ccb", + "d9dfe4578bc891fe65984d0daaad1e9783ac51de" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_2500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0_small", + "source_revision": "05586f38706727534a0f388ae0ece46541776d65", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_0_small/checkpoints/steps_1500_model.safetensors", + "identity": "bd2796b952df7654f0376b226923307644f24d90232731083919241d2a269f62", + "runpath": "historical/openvla_flappy_fix_latency_0_small", + "configpath": "historical/openvla_flappy_fix_latency_0_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_0_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_0_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_1500_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7", + "source_revision": "94014d226e26fcf46f117f6bea42a59b0be2d753", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7/checkpoints/steps_4000_model.safetensors", + "identity": "c3bb85e1e63acae611867af90a6d1ae221dcd03b30d54f0f3ed0175c2a9824b1", + "runpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7", + "configpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7/config.yaml", + "statpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7/dataset_statistics.json", + "run_id": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_4000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "demon_attack_fix_latency_2_200ep/checkpoints/steps_7000_model.safetensors", + "identity": "2ab6d4d4eb795a9a09da9c63d489771039440689877642e3f9ec48dd9112119a", + "runpath": "demon_attack_fix_latency_2_200ep", + "configpath": "demon_attack_fix_latency_2_200ep/config.yaml", + "statpath": "demon_attack_fix_latency_2_200ep/dataset_statistics.json", + "run_id": "demon_attack_fix_latency_2_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + ], + "loader_revision": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_7000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "demon_attack_fix_latency_6_200ep/checkpoints/steps_7000_model.safetensors", + "identity": "375ff9f24479e219e8197bbc2eee3e71e61e1853e0c80eefe438ca947d5c5b1b", + "runpath": "demon_attack_fix_latency_6_200ep", + "configpath": "demon_attack_fix_latency_6_200ep/config.yaml", + "statpath": "demon_attack_fix_latency_6_200ep/dataset_statistics.json", + "run_id": "demon_attack_fix_latency_6_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + ], + "loader_revision": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_7000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/dfb71eef10b42a14f04f09b297665fcbb6cf77bd", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3", + "source_revision": "1967aac3fadf905a7f693e388cbf2aebdd40e5de", + "repo": "latency-sensitive-bench/latency-information-models", + "revision": "b19c1eca8caefeaafcfbfc39f53c240c4e9a255c", + "path": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3/checkpoints/steps_5000_pytorch_model.pt", + "identity": "2f41824572db70b0517fffee698dfbd6a2274494eb6aecac82e6f91edf47a219", + "runpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3", + "configpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3/config.yaml", + "statpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3/dataset_statistics.json", + "run_id": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "talha1503", + "wandb_project": "latency-sensitive-bench", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_3_small", + "source_revision": "86b4d314c377420fa8eb9ac7aa268695c2ab4f40", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_3_small/checkpoints/best_model.safetensors", + "identity": "dfd6f77b285dff5a628b8a2e8aee12d2c8f11c2196464ed2505f7e138ec2a828", + "runpath": "historical/openvla_flappy_fix_latency_3_small", + "configpath": "historical/openvla_flappy_fix_latency_3_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_3_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_3_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small", + "source_revision": "7842a04e6ba8ed4dc56ccdae235735a807a77b5d", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_1_small/checkpoints/best_model.safetensors", + "identity": "758b4ccf870860c583af4da492b2d6782333e847759afd7314ba4547610fb04e", + "runpath": "historical/openvla_flappy_fix_latency_1_small", + "configpath": "historical/openvla_flappy_fix_latency_1_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_1_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_1_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small_tmp", + "source_revision": "8fba8c350f6f10d70119807a546218cfa82e9150", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_1_small_tmp/checkpoints/best_model.safetensors", + "identity": "6abc545e6b1abb272b7829b5676f412eebf33cfe0f21ba56368eb619c4610d83", + "runpath": "historical/openvla_flappy_fix_latency_1_small_tmp", + "configpath": "historical/openvla_flappy_fix_latency_1_small_tmp/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_1_small_tmp/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_1", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "0bc71a07a07274d35bc9aa60f8d8d14306d45c24", + "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_0_small", + "source_revision": "aeefbbd9676d95973aea2a6f63c433807a44d05b", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_demon_attack_fix_latency_0_small/checkpoints/best_model.safetensors", + "identity": "0eda74a2340c7b2df580ca0589505c47ba32f7244b54a15878f61f86fa922213", + "runpath": "historical/openvla_demon_attack_fix_latency_0_small", + "configpath": "historical/openvla_demon_attack_fix_latency_0_small/config.yaml", + "statpath": "historical/openvla_demon_attack_fix_latency_0_small/dataset_statistics.json", + "run_id": "openvla_demon_attack_fix_latency_0", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "917d564a90c8209144f9b50204c47192f27e1ac1", + "d9dfe4578bc891fe65984d0daaad1e9783ac51de" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_small", + "source_revision": "371074f07545d1d4e5533afc4d457a63d03bf9da", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_4_small/checkpoints/best_model.safetensors", + "identity": "f893f7feb006ca5eee1dbc84ac222cf50c7b0dddcb7d572cf6d52135605bb0b9", + "runpath": "historical/openvla_flappy_fix_latency_4_small", + "configpath": "historical/openvla_flappy_fix_latency_4_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_4_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_4_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + ], + "loader_revision": "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2_small", + "source_revision": "87c38afed1a227e92aac9efeb0921f57d556df8a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_2_small/checkpoints/best_model.safetensors", + "identity": "ee25ed102ecbf335b536f8260de7ae96b84cfa7158024c19e6dc51d32cd9ed27", + "runpath": "historical/openvla_flappy_fix_latency_2_small", + "configpath": "historical/openvla_flappy_fix_latency_2_small/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_2_small/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_2_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "d182480d774978c8e37efb1d3ea983d812acd4f6" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2", + "source_revision": "8135db54520527abe4a62f002323a307d1a47c2d", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_2/checkpoints/steps_5000_pytorch_model.pt", + "identity": "7568ff3e7f09a67fd0681287750fbe5c34c99617b08bb440997c97b247c5d651", + "runpath": "historical/openvla_flappy_fix_latency_2", + "configpath": "historical/openvla_flappy_fix_latency_2/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_2/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1e638dd5cb934c2e322f2f3879174c7d626e4410", + "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_2_small", + "source_revision": "bb91ae40a1bb9b9f6a93be031193de603da86ecd", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_demon_attack_fix_latency_2_small/checkpoints/best_model.safetensors", + "identity": "adb853d16890002501675633ddefc2126fd3f102a63cf6ece7a5e56d0c8c662c", + "runpath": "historical/openvla_demon_attack_fix_latency_2_small", + "configpath": "historical/openvla_demon_attack_fix_latency_2_small/config.yaml", + "statpath": "historical/openvla_demon_attack_fix_latency_2_small/dataset_statistics.json", + "run_id": "openvla_demon_attack_fix_latency_2_small", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2", + "source_revision": "c4cbca936a122c2d55b1edee3959656f8174c5b7", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2/checkpoints/steps_5000_model.safetensors", + "identity": "2506bd51c7054d65cc980738979dd484de07b4af15da554185cbc28ede512de5", + "runpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2", + "configpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2/config.yaml", + "statpath": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2/dataset_statistics.json", + "run_id": "openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2", + "source_revision": "177c11a86ee64432264ae8f1e577ecfd6c767918", + "repo": "latency-sensitive-bench/latency-information-models", + "revision": "b19c1eca8caefeaafcfbfc39f53c240c4e9a255c", + "path": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2/checkpoints/steps_5000_pytorch_model.pt", + "identity": "28b6ca56206555b081f481247649ac06762debcb7cf74776b2ca240338d14974", + "runpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2", + "configpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2/config.yaml", + "statpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2/dataset_statistics.json", + "run_id": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "talha1503", + "wandb_project": "latency-sensitive-bench", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "9915dcb0dc644aa2ed14172f298386f53b03c6f6", + "repo": "latency-sensitive-bench/memory-models", + "revision": "555189de648aa787fd65c8413d0459eac98232bf", + "path": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15/checkpoints/steps_5000_model.safetensors", + "identity": "91931822871b9c3af55f58a7254963b512e478bda6d7514dd2a194ac27311cd5", + "runpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "configpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15/config.yaml", + "statpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15/dataset_statistics.json", + "run_id": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_l2_lora", + "source_revision": "427f4c0582d6eb9419fa39f5942db1d59585993c", + "repo": "latency-sensitive-bench/vla-finetuning-scope-models", + "revision": "c1275f13f74269e997f8f369d164931279f9014a", + "path": "historical/openvla_flappy_l2_lora/final_model/adapter_model.safetensors", + "identity": "551bd2362cc9548b9d97f3a4f77fcfbe2b28f2725b2897ab42b996485962b8bb", + "runpath": "historical/openvla_flappy_l2_lora", + "configpath": "historical/openvla_flappy_l2_lora/config.yaml", + "statpath": "historical/openvla_flappy_l2_lora/dataset_statistics.json", + "run_id": "openvla_flappy_l2_lora", + "kind": "lora", + "native_ready": true, + "base_model": "action", + "framework": "QwenOFT", + "wandb_entity": "saberrr-zju", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "9aa8c96e8044e6dc48063a9f7e0b13df513ebf62" + ], + "loader_revision": "9aa8c96e8044e6dc48063a9f7e0b13df513ebf62", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/final_model", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/9aa8c96e8044e6dc48063a9f7e0b13df513ebf62", + "backbone": "/tmp/experiment-repo-validation-20261007/action-base-1/models--StarVLA--Qwen3-VL-4B-Instruct-Action/snapshots/c41500cf1d287cd79e9f6602f419937b05048bd8", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 1442, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_exp", + "source_revision": "213bab6c06b13c73fee875412d99e40ea4a4dd70", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_4_exp/checkpoints/steps_2500_pytorch_model.pt", + "identity": "1610fb6615308c849e844acde8d98b233fe6f6c4d01657d184fd4bd9945e55f5", + "runpath": "historical/openvla_flappy_fix_latency_4_exp", + "configpath": "historical/openvla_flappy_fix_latency_4_exp/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_4_exp/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_4_exp", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "46e36dafd933fd767313c1c84f7ace66c2683be2" + ], + "loader_revision": "46e36dafd933fd767313c1c84f7ace66c2683be2", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_2500_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/46e36dafd933fd767313c1c84f7ace66c2683be2", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "flappy_curriculum_cumulative_40ep_per_latency/checkpoints/steps_5000_model.safetensors", + "identity": "ad4903982cf1ec6f29c8ae4edea25e10b42353d56e0755b5ed123f5df3351754", + "runpath": "flappy_curriculum_cumulative_40ep_per_latency", + "configpath": "flappy_curriculum_cumulative_40ep_per_latency/config.yaml", + "statpath": "flappy_curriculum_cumulative_40ep_per_latency/dataset_statistics.json", + "run_id": "flappy_curriculum_cumulative_40ep_per_latency", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "d1a786ff88f4a5aff08baa76b8c82ab4964a2804" + ], + "loader_revision": "d1a786ff88f4a5aff08baa76b8c82ab4964a2804", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/d1a786ff88f4a5aff08baa76b8c82ab4964a2804", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "flappy_fix_latency_0_200ep/checkpoints/steps_5000_model.safetensors", + "identity": "7e22c0b5c6118ff5bac5536d076ef7c8a51cf14cfa8b86752464321ac7311d39", + "runpath": "flappy_fix_latency_0_200ep", + "configpath": "flappy_fix_latency_0_200ep/config.yaml", + "statpath": "flappy_fix_latency_0_200ep/dataset_statistics.json", + "run_id": "flappy_fix_latency_0_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "fa2cf48737a4e7c4942bb9ac8d6d2e52880fb456" + ], + "loader_revision": "fa2cf48737a4e7c4942bb9ac8d6d2e52880fb456", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/fa2cf48737a4e7c4942bb9ac8d6d2e52880fb456", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "flappy_fix_latency_2_200ep/checkpoints/steps_5000_model.safetensors", + "identity": "aabcc8e3a22947e22d161d324de0e613a03300f39fa1dd28bffa2df927e994cf", + "runpath": "flappy_fix_latency_2_200ep", + "configpath": "flappy_fix_latency_2_200ep/config.yaml", + "statpath": "flappy_fix_latency_2_200ep/dataset_statistics.json", + "run_id": "flappy_fix_latency_2_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "ffe48f30f636bd6984f9d8c37e9b54c9f474c172" + ], + "loader_revision": "ffe48f30f636bd6984f9d8c37e9b54c9f474c172", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/ffe48f30f636bd6984f9d8c37e9b54c9f474c172", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "flappy_fix_latency_4_200ep/checkpoints/steps_5000_model.safetensors", + "identity": "99f4d57787efbfd68e5fc68789f72e7e77060c0061568006a8ce8c721c73b305", + "runpath": "flappy_fix_latency_4_200ep", + "configpath": "flappy_fix_latency_4_200ep/config.yaml", + "statpath": "flappy_fix_latency_4_200ep/dataset_statistics.json", + "run_id": "flappy_fix_latency_4_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "ffe48f30f636bd6984f9d8c37e9b54c9f474c172" + ], + "loader_revision": "ffe48f30f636bd6984f9d8c37e9b54c9f474c172", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/ffe48f30f636bd6984f9d8c37e9b54c9f474c172", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0", + "source_revision": "06f64d1a03905ad0f04a447c57afb970c29a559e", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_0/checkpoints/steps_5000_model.safetensors", + "identity": "57b21a1d51a95495f08c60166b561aed0183fd4b680b665a440b9b0694aff4c8", + "runpath": "historical/openvla_flappy_fix_latency_0", + "configpath": "historical/openvla_flappy_fix_latency_0/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_0/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_0", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "your_wandb_entity", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0", + "source_revision": "9f8c6842151405a82a3e852d2a00e27885db1e93", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_deadly_corridor_fix_latency_0/checkpoints/best_model.safetensors", + "identity": "1f2b7941653b956dbb5c368b9da60e5538e04216c34371254ef6cfa5c34bce74", + "runpath": "historical/openvla_deadly_corridor_fix_latency_0", + "configpath": "historical/openvla_deadly_corridor_fix_latency_0/config.yaml", + "statpath": "historical/openvla_deadly_corridor_fix_latency_0/dataset_statistics.json", + "run_id": "openvla_deadly_corridor_fix_latency_0", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1b57b562a6c2ffb2fd549282e0baaa1f86661123", + "914436f584ae969603a7e9b617a400f266e7fb9f", + "917d564a90c8209144f9b50204c47192f27e1ac1", + "9e1cf489a76ed2c07abc29399d637d5d73417ccb", + "d9dfe4578bc891fe65984d0daaad1e9783ac51de" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/best_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_fix_latency_6", + "source_revision": "7a2e2f81dda9fc9de068fa15b1eac442ed6fb51f", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "3a3aac10736c67f1a3fdb0f4f2bac0348d9c6ef6", + "path": "historical/openvla_flappy_fix_latency_6/checkpoints/steps_5000_model.safetensors", + "identity": "a24aac79d3d3fd72e0a8e782c0fd1f5b3294ee74eca4bbce1cbff73128b72122", + "runpath": "historical/openvla_flappy_fix_latency_6", + "configpath": "historical/openvla_flappy_fix_latency_6/config.yaml", + "statpath": "historical/openvla_flappy_fix_latency_6/dataset_statistics.json", + "run_id": "openvla_flappy_fix_latency_6", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1e638dd5cb934c2e322f2f3879174c7d626e4410" + ], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "compatibility_contract": "Documented Qwen tied LM-head checkpoint loading", + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + } +] \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/old_code_recovery.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/old_code_recovery.json new file mode 100644 index 0000000000000000000000000000000000000000..48e9bcf2ec3e6e94a97f9651eff65cfd5d1898f3 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/old_code_recovery.json @@ -0,0 +1,10 @@ +[ + { + "commit": "538fd523fe4fa4e7abce3201611919314bc6b664", + "http_status": 200 + }, + { + "commit": "9aa8c96e8044e6dc48063a9f7e0b13df513ebf62", + "http_status": 200 + } +] \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/original_reference_documents/latency-transfer-data/run_index.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/original_reference_documents/latency-transfer-data/run_index.json new file mode 100644 index 0000000000000000000000000000000000000000..c8d472dd8f7cf522aed2509686e576b23e27b666 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/original_reference_documents/latency-transfer-data/run_index.json @@ -0,0 +1,144 @@ +[ + { + "run": "wan_oft_flappy_fix_latency_0_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 0, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-06-28T13:44:26.129154Z", + "git_commit": "f9acb69ef3d99d2a8b620103161e15eebc4c3091", + "wandb_run_id": "ky48y0be", + "checkpoint": "wan_oft_flappy_fix_latency_0_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "daa8a61baa149139792808dcfc5406df766d39113161b92d0baa13a9bd2402f4", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_0_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 279882, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_1_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 1, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-06-29T14:50:14.198406Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "3kohuetr", + "checkpoint": "wan_oft_flappy_fix_latency_1_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "ca01daf3e577fedcc238882e531c748f6a5c178934ca7cc7209da18eb24ff03f", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_1_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 297483, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_2_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 2, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T09:28:56.645196Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "2z7am65s", + "checkpoint": "wan_oft_flappy_fix_latency_2_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "328140abd4682a66cb56cdbe03524f6c7af534e2c9ac2bd23ac1c9691363957a", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_2_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 330734, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid", + "fig13_wanoft_timing_controls", + "fig14_wanoft_timing_buckets" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_3_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 3, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T16:06:06.153675Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "jlutvxqx", + "checkpoint": "wan_oft_flappy_fix_latency_3_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "d5940d8ea363b40611f391103534da4d13dd7204e43024079c56eb2f34322bcc", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_3_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 316258, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_4_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 4, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T17:12:40.334458Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "hvokzu73", + "checkpoint": "wan_oft_flappy_fix_latency_4_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "05a1c777b84de4f2b66a2db1a13eb7b1aa326167eabb30ddf50bd149d6e68118", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_4_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 311588, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + } +] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/original_reference_documents/latency-transfer-models/run_index.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/original_reference_documents/latency-transfer-models/run_index.json new file mode 100644 index 0000000000000000000000000000000000000000..545ab189075410ffb205df39387829e0f16d01e7 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/original_reference_documents/latency-transfer-models/run_index.json @@ -0,0 +1,144 @@ +[ + { + "run": "wan_oft_flappy_fix_latency_0_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 0, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-06-28T13:44:26.129154Z", + "git_commit": "f9acb69ef3d99d2a8b620103161e15eebc4c3091", + "wandb_run_id": "ky48y0be", + "checkpoint": "wan_oft_flappy_fix_latency_0_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "daa8a61baa149139792808dcfc5406df766d39113161b92d0baa13a9bd2402f4", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "23f4501059d58f7df4d12d7a074f48daca4c578d", + "path": "training/flappy_fix_latency_0_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 279882, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_1_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 1, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-06-29T14:50:14.198406Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "3kohuetr", + "checkpoint": "wan_oft_flappy_fix_latency_1_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "ca01daf3e577fedcc238882e531c748f6a5c178934ca7cc7209da18eb24ff03f", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "23f4501059d58f7df4d12d7a074f48daca4c578d", + "path": "training/flappy_fix_latency_1_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 297483, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_2_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 2, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T09:28:56.645196Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "2z7am65s", + "checkpoint": "wan_oft_flappy_fix_latency_2_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "328140abd4682a66cb56cdbe03524f6c7af534e2c9ac2bd23ac1c9691363957a", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "23f4501059d58f7df4d12d7a074f48daca4c578d", + "path": "training/flappy_fix_latency_2_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 330734, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid", + "fig13_wanoft_timing_controls", + "fig14_wanoft_timing_buckets" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_3_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 3, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T16:06:06.153675Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "jlutvxqx", + "checkpoint": "wan_oft_flappy_fix_latency_3_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "d5940d8ea363b40611f391103534da4d13dd7204e43024079c56eb2f34322bcc", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "23f4501059d58f7df4d12d7a074f48daca4c578d", + "path": "training/flappy_fix_latency_3_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 316258, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_4_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 4, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T17:12:40.334458Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "hvokzu73", + "checkpoint": "wan_oft_flappy_fix_latency_4_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "05a1c777b84de4f2b66a2db1a13eb7b1aa326167eabb30ddf50bd149d6e68118", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "23f4501059d58f7df4d12d7a074f48daca4c578d", + "path": "training/flappy_fix_latency_4_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 311588, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + } +] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/pinned_setup_tests.log b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/pinned_setup_tests.log new file mode 100644 index 0000000000000000000000000000000000000000..33b2fbe6e8da16762b0f27b17dcc149aa56af5c3 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/pinned_setup_tests.log @@ -0,0 +1,17 @@ +........ [100%] +=============================== warnings summary =============================== +third_party/starVLA/tests/rl_games/test_pi05_flappy_sft_path.py::test_launch_train_setup_namespace_uses_composed_hydra_config +third_party/starVLA/tests/rl_games/test_pi05_flappy_sft_path.py::test_launch_train_setup_namespace_forwards_raw_frame_latency_unit +third_party/starVLA/tests/rl_games/test_pi05_flappy_sft_path.py::test_launch_train_setup_namespace_forwards_deadly_factorized_layout +third_party/starVLA/tests/rl_games/test_pi05_flappy_sft_path.py::test_launch_train_setup_namespace_forwards_explicit_dataset_source_hf +third_party/starVLA/tests/rl_games/test_pi05_flappy_sft_path.py::test_launch_train_setup_namespace_forwards_dataset_config_name +third_party/starVLA/tests/rl_games/test_pi05_flappy_sft_path.py::test_launch_train_setup_namespace_forwards_dataset_source_subdir + /home/lixinyuan/latency-sensitive-bench/third_party/starVLA/examples/rl_games/scripts/launch_train.py:278: Hydra14MigrationWarning: + version_base="1.1" selects Hydra 1.1 compatibility behavior. + Hydra 1.4 will remove this compatibility behavior. + Before upgrading to Hydra 1.4, set version_base="1.3" and test your application. + See https://hydra.cc/docs/upgrades/1.3_to_1.4/prepare_for_1_4/ for preparation instructions. + with initialize_config_dir(version_base="1.1", config_dir=str(CONFIG_DIR)): + +-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html +8 passed, 32 deselected, 6 warnings in 1.32s diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_artifact_contents.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_artifact_contents.json new file mode 100644 index 0000000000000000000000000000000000000000..14616a0bf7084ce0edb706f34a139155edf46a99 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_artifact_contents.json @@ -0,0 +1,62 @@ +{ + "data": { + "project": { + "run": { + "outputArtifacts": { + "edges": [ + { + "node": { + "artifactSequence": { + "name": "run-i4aobzgd-comparisonspre_post-7S3BqA" + }, + "versionIndex": 0, + "fileCount": 1, + "size": 6676 + } + }, + { + "node": { + "artifactSequence": { + "name": "run-i4aobzgd-comparisonscontrol_drops-RhD2_Q" + }, + "versionIndex": 0, + "fileCount": 1, + "size": 10945 + } + }, + { + "node": { + "artifactSequence": { + "name": "wan_oft_flappy_latency2_context5_temporal_linear_probe_v0-report" + }, + "versionIndex": 0, + "fileCount": 1, + "size": 346819 + } + }, + { + "node": { + "artifactSequence": { + "name": "run-i4aobzgd-events" + }, + "versionIndex": 0, + "fileCount": 1, + "size": 462655 + } + }, + { + "node": { + "artifactSequence": { + "name": "run-i4aobzgd-history" + }, + "versionIndex": 0, + "fileCount": 1, + "size": 142005 + } + } + ] + } + } + } + } +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_artifacts.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_artifacts.json new file mode 100644 index 0000000000000000000000000000000000000000..7e2c6b09a199daacad978b4c94fd42dc81188599 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_artifacts.json @@ -0,0 +1,174 @@ +{ + "data": { + "__type": { + "fields": [ + { + "name": "aliases" + }, + { + "name": "artifactCollections" + }, + { + "name": "artifactLineageDag" + }, + { + "name": "artifactMemberships" + }, + { + "name": "artifactSequence" + }, + { + "name": "artifactSequenceMembership" + }, + { + "name": "artifactType" + }, + { + "name": "commitHash" + }, + { + "name": "createdAt" + }, + { + "name": "createdBy" + }, + { + "name": "currentManifest" + }, + { + "name": "dependencyOf" + }, + { + "name": "dependsOn" + }, + { + "name": "description" + }, + { + "name": "digest" + }, + { + "name": "digestAlgorithm" + }, + { + "name": "fileCount" + }, + { + "name": "historyStep" + }, + { + "name": "id" + }, + { + "name": "isGenerated" + }, + { + "name": "isLinkedToGlobalRegistry" + }, + { + "name": "isProtected" + }, + { + "name": "metadata" + }, + { + "name": "runHistory" + }, + { + "name": "size" + }, + { + "name": "state" + }, + { + "name": "storageBytes" + }, + { + "name": "storageRegion" + }, + { + "name": "tags" + }, + { + "name": "ttlDurationSeconds" + }, + { + "name": "ttlIsInherited" + }, + { + "name": "updatedAt" + }, + { + "name": "usedBy" + }, + { + "name": "usedCount" + }, + { + "name": "versionIndex" + }, + { + "name": "weaveSource" + } + ] + }, + "project": { + "run": { + "outputArtifacts": { + "edges": [ + { + "node": { + "id": "QXJ0aWZhY3Q6MzEzNTYwOTQ4OQ==", + "digest": "f4a668adb863ad8605b253e099858c14", + "artifactSequence": { + "name": "run-i4aobzgd-comparisonspre_post-7S3BqA" + }, + "versionIndex": 0 + } + }, + { + "node": { + "id": "QXJ0aWZhY3Q6MzEzNTYwOTUwMw==", + "digest": "4833225fa4fb4d9793cd8a9006b7b51b", + "artifactSequence": { + "name": "run-i4aobzgd-comparisonscontrol_drops-RhD2_Q" + }, + "versionIndex": 0 + } + }, + { + "node": { + "id": "QXJ0aWZhY3Q6MzEzNTYwOTUxMA==", + "digest": "4c586288f98a7faefd0f4083b37c2475", + "artifactSequence": { + "name": "wan_oft_flappy_latency2_context5_temporal_linear_probe_v0-report" + }, + "versionIndex": 0 + } + }, + { + "node": { + "id": "QXJ0aWZhY3Q6MzEzNTY1NDM5Mg==", + "digest": "571adb4e5df7e175ebe4b63a5e24cd65", + "artifactSequence": { + "name": "run-i4aobzgd-events" + }, + "versionIndex": 0 + } + }, + { + "node": { + "id": "QXJ0aWZhY3Q6MzEzNTY1NDY0Ng==", + "digest": "3290c1666a63c7b0ccc45e40ef84e328", + "artifactSequence": { + "name": "run-i4aobzgd-history" + }, + "versionIndex": 0 + } + } + ] + } + } + } + } +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_files.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_files.json new file mode 100644 index 0000000000000000000000000000000000000000..f0281fe23f995a0d60f48d3d4df0f3dc9940395b --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/probe_wandb_files.json @@ -0,0 +1,97 @@ +{ + "data": { + "project": { + "run": { + "id": "UnVuOnYxOmk0YW9iemdkOnN0YXJWTEFfcmxfZ2FtZXM6ZG9uZ3FpYW55dTk5LXpoZWppYW5nLXVuaXZlcnNpdHk=", + "files": { + "edges": [ + { + "node": { + "name": "artifact/3135609489/wandb_manifest.json", + "sizeBytes": 335, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/artifact/3135609489/wandb_manifest.json" + } + }, + { + "node": { + "name": "artifact/3135609503/wandb_manifest.json", + "sizeBytes": 341, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/artifact/3135609503/wandb_manifest.json" + } + }, + { + "node": { + "name": "artifact/3135609510/wandb_manifest.json", + "sizeBytes": 317, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/artifact/3135609510/wandb_manifest.json" + } + }, + { + "node": { + "name": "artifact/3135654392/wandb_manifest.json", + "sizeBytes": 0, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/artifact/3135654392/wandb_manifest.json" + } + }, + { + "node": { + "name": "artifact/3135654646/wandb_manifest.json", + "sizeBytes": 0, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/artifact/3135654646/wandb_manifest.json" + } + }, + { + "node": { + "name": "config.yaml", + "sizeBytes": 6146, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/config.yaml" + } + }, + { + "node": { + "name": "media/table/comparisons/control_drops_7_bc05a2ab4fb5dbc07ef5.table.json", + "sizeBytes": 9500, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/media/table/comparisons/control_drops_7_bc05a2ab4fb5dbc07ef5.table.json" + } + }, + { + "node": { + "name": "media/table/comparisons/pre_post_7_cb9c105bb875bb816493.table.json", + "sizeBytes": 5562, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/media/table/comparisons/pre_post_7_cb9c105bb875bb816493.table.json" + } + }, + { + "node": { + "name": "output.log", + "sizeBytes": 2405, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/output.log" + } + }, + { + "node": { + "name": "requirements.txt", + "sizeBytes": 3120, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/requirements.txt" + } + }, + { + "node": { + "name": "wandb-metadata.json", + "sizeBytes": 2993, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/wandb-metadata.json" + } + }, + { + "node": { + "name": "wandb-summary.json", + "sizeBytes": 52580, + "url": "https://api.wandb.ai/files/dongqianyu99-zhejiang-university/starVLA_rl_games/i4aobzgd/wandb-summary.json" + } + } + ] + } + } + } + } +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/source_inventory.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/source_inventory.json new file mode 100644 index 0000000000000000000000000000000000000000..de88cef7eeb902dd348e68775771ee8d8f7ba129 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/source_inventory.json @@ -0,0 +1,642 @@ +[ + { + "batch": "0", + "type": "model", + "repo": "latency-sensitive-bench/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce", + "revision": "8cf964f252337912a8f8e5d30467a33f4e8a49dc", + "files": 17, + "bytes": 23474220474 + }, + { + "batch": "0", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_0_200ep_context5", + "revision": "f5f82fb2c760bef7ddb8882dbba18075fa1cf2c8", + "files": 5, + "bytes": 10602052692 + }, + { + "batch": "0", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_1_200ep_context5", + "revision": "8ee61cea50b7c3fcad97f4ab2f41f97044d219a8", + "files": 5, + "bytes": 10841997219 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0", + "revision": "06f64d1a03905ad0f04a447c57afb970c29a559e", + "files": 12, + "bytes": 18277031882 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_6", + "revision": "7a2e2f81dda9fc9de068fa15b1eac442ed6fb51f", + "files": 12, + "bytes": 18277056314 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2", + "revision": "8135db54520527abe4a62f002323a307d1a47c2d", + "files": 12, + "bytes": 18277063818 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_exp", + "revision": "213bab6c06b13c73fee875412d99e40ea4a4dd70", + "files": 12, + "bytes": 18276903161 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_small", + "revision": "371074f07545d1d4e5533afc4d457a63d03bf9da", + "files": 13, + "bytes": 18276621481 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small_tmp", + "revision": "8fba8c350f6f10d70119807a546218cfa82e9150", + "files": 13, + "bytes": 18276635611 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0_small", + "revision": "05586f38706727534a0f388ae0ece46541776d65", + "files": 13, + "bytes": 18276617707 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small", + "revision": "7842a04e6ba8ed4dc56ccdae235735a807a77b5d", + "files": 13, + "bytes": 18276610436 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2_small", + "revision": "87c38afed1a227e92aac9efeb0921f57d556df8a", + "files": 13, + "bytes": 18276621420 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_3_small", + "revision": "86b4d314c377420fa8eb9ac7aa268695c2ab4f40", + "files": 13, + "bytes": 18276629117 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0_full", + "revision": "9281fd061894c14d3641ee9724e108cd33baf494", + "files": 13, + "bytes": 18276748150 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0", + "revision": "9f8c6842151405a82a3e852d2a00e27885db1e93", + "files": 13, + "bytes": 18276733726 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_2", + "revision": "0a60d6c8a400b48e4d9737b36ef3a44814d52d2a", + "files": 13, + "bytes": 18276730845 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_0_small", + "revision": "aeefbbd9676d95973aea2a6f63c433807a44d05b", + "files": 11, + "bytes": 9138432714 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_2_small", + "revision": "bb91ae40a1bb9b9f6a93be031193de603da86ecd", + "files": 13, + "bytes": 18276668819 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_6_small", + "revision": "9e8e2e6cc073eb68bd59306ab1141972ed1ed819", + "files": 1, + "bytes": 1519 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_200ep", + "revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "files": 71, + "bytes": 63981635137 + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "files": 51, + "bytes": 45705520630 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_2", + "revision": "c0605fa1df28b19249849c3e748bbc445bef6526", + "files": 5, + "bytes": 257552608 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_1", + "revision": "7190bf45dfface92ec7cbaa3c774a7e29616db52", + "files": 5, + "bytes": 256670462 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_0", + "revision": "2deacb435543c52ae51f3c947bf917fdb0b5493c", + "files": 5, + "bytes": 254545643 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_3", + "revision": "709111a3b59e413214e0417bd878291734ec9427", + "files": 5, + "bytes": 251773874 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_4", + "revision": "c7f6eba84f512bc22a095971c9b75499e841a026", + "files": 5, + "bytes": 234519256 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_5", + "revision": "67876ea9a8d5c687b924199574e298c68bf9b83d", + "files": 5, + "bytes": 232364317 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_7", + "revision": "d76a57ef1b84c591c47975347c539da59670f3b4", + "files": 5, + "bytes": 205595492 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_6", + "revision": "0766ba4ea9196b861302e99e564b7d2e05453885", + "files": 5, + "bytes": 200941670 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_8", + "revision": "e02c6b1db39b37965569d4f559f1d710fb55c58a", + "files": 5, + "bytes": 147343953 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_12", + "revision": "ac9d21a59b908e717de2bee2cc749b90495a00b8", + "files": 5, + "bytes": 147328521 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_9", + "revision": "7be93d38340141e385913e540b028024d74f1cf8", + "files": 5, + "bytes": 140841771 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_13", + "revision": "2a5c5d2aa4730db22fd33f4f00f4d93fc5043465", + "files": 5, + "bytes": 133539812 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_14", + "revision": "9867d03ad02a4d025cb6cc88281f95596f83a814", + "files": 5, + "bytes": 128938837 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_10", + "revision": "9a3f7f736402b063885068e683860b6746e5337d", + "files": 5, + "bytes": 128832856 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_15", + "revision": "21811c6039f0a538f11821730ee6792bd0d6c0b4", + "files": 5, + "bytes": 110909215 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_11", + "revision": "78e4e1b6d9bb57c5f30712cd0ed2f3b02877ff56", + "files": 5, + "bytes": 91939207 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_0", + "revision": "6ee54a31fe5645bce299adeab17a68ff458bfabe", + "files": 5, + "bytes": 67749647 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_12", + "revision": "258c4ae285ad34445578dea4e786c408fcd6be30", + "files": 5, + "bytes": 44793589 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_4", + "revision": "4dee6c0fd4abb86507001c4dc7404a5310c2f30a", + "files": 5, + "bytes": 43805390 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_13", + "revision": "dbb7678df6396735290099d070dfb5827a4cf4a4", + "files": 5, + "bytes": 43616311 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_2", + "revision": "df04b6f30856b2903e65a6a870f7440055f02311", + "files": 5, + "bytes": 42670186 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_14", + "revision": "fb78a1bc6280a4d0a6a60e8231d63d848819bd1e", + "files": 5, + "bytes": 40543661 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_5", + "revision": "f614a91258d86cb34cd2e3e0e79e2bf69b321e76", + "files": 5, + "bytes": 40279413 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_7", + "revision": "d04b9bc42bfbb7da400438a9fb75c49b198f7776", + "files": 5, + "bytes": 38892941 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_1", + "revision": "e09b7f2c51726bb169d827fe973b2d429b6b1432", + "files": 5, + "bytes": 36658578 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_10", + "revision": "94762ebd79edb55d6e83ba2046729e3e6574d64d", + "files": 5, + "bytes": 36560892 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_15", + "revision": "c586ddd84f76b6295de7929b62ad4da3359b0348", + "files": 5, + "bytes": 36311418 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_6", + "revision": "47cfced1713ce7d57d4567110269fb5929e8e7a8", + "files": 5, + "bytes": 36072348 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_9", + "revision": "3405ac7c12ca266e2d3e83047bbd06e1e5eed220", + "files": 5, + "bytes": 32290799 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_3", + "revision": "b8b18213fe10aa0a6be2b7602e23c2ac938b49e0", + "files": 5, + "bytes": 31980248 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_8", + "revision": "164f4e56abb6c2637801e779918a99a7d70c9c3a", + "files": 5, + "bytes": 29960490 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_3_small", + "revision": "8470ae1f00ca10bb8646ac7cfdccf9682e053eab", + "files": 7, + "bytes": 1272008595 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_2_small", + "revision": "c9bbd1744020a587b55590c897628510e0741136", + "files": 7, + "bytes": 1271812775 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_1_small", + "revision": "749a849911b8c35005c26241fb4edbcee1286832", + "files": 7, + "bytes": 1266682400 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_0_small", + "revision": "7e974f1e8e1ad7ca4ee0ffad0849d886d64128d3", + "files": 6, + "bytes": 1230559183 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_4_small", + "revision": "8a6a091570530476b313976b6a8d1b94d600f6d9", + "files": 6, + "bytes": 1211322894 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_200ep", + "revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "files": 22, + "bytes": 21680187271 + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_200ep", + "revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "files": 21, + "bytes": 1148999664 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "revision": "5dfd90884fc4e3817bc51f574d6fc9cea2d7948b", + "files": 8, + "bytes": 9784919229 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3", + "revision": "f8d64aff21628adbdd0d1aaddd1bc7067bc2a23c", + "files": 17, + "bytes": 9784948330 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "revision": "9915dcb0dc644aa2ed14172f298386f53b03c6f6", + "files": 8, + "bytes": 9138248018 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "revision": "cd810024c06f961c95449785f19ab73ccfd20de1", + "files": 13, + "bytes": 9138255732 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2", + "revision": "73ea97c064d206c3a0dcf6ffdbf2ecfe97bc2c3c", + "files": 8, + "bytes": 9138247431 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2", + "revision": "c4cbca936a122c2d55b1edee3959656f8174c5b7", + "files": 13, + "bytes": 9138255777 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1", + "revision": "004ec237af79c10de5c0d4376cf8a13542ac01f8", + "files": 8, + "bytes": 9138247433 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1", + "revision": "bc733d64e32512a8fc76f9be022eb5b0d7c0e2cb", + "files": 13, + "bytes": 9138255750 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7", + "revision": "94014d226e26fcf46f117f6bea42a59b0be2d753", + "files": 8, + "bytes": 9138247812 + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2", + "revision": "ac5ff1109c46295299d74b5e30a56f1e27e930a6", + "files": 5, + "bytes": 14628 + }, + { + "batch": "2", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "revision": "f90d1dac8b31a18c0496884729c744d7a4c877ae", + "files": 4, + "bytes": 25946174934 + }, + { + "batch": "2", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "revision": "862b6cad9b2c87dc129d5cf45a9b23b18450c4ef", + "files": 6, + "bytes": 993634330 + }, + { + "batch": "3", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_l2_full", + "revision": "2e8e58c6af1ee6f8a3701388b2552a4fa65a041c", + "files": 6, + "bytes": 9805979462 + }, + { + "batch": "3", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_l2_lora", + "revision": "427f4c0582d6eb9419fa39f5942db1d59585993c", + "files": 9, + "bytes": 292406543 + }, + { + "batch": "3", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fixed_l2_fs4", + "revision": "1bdd2ea6ea7db07139e4694f445c12198399c2b8", + "files": 5, + "bytes": 11243685412 + }, + { + "batch": "3", + "type": "dataset", + "repo": "latency-sensitive-bench/lora_comparison_flappy_l2_eval", + "revision": "97db426f51a9427fbe1e8c005b33dbe7aeac3899", + "files": 6, + "bytes": 113503 + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2", + "revision": "177c11a86ee64432264ae8f1e577ecfd6c767918", + "files": 40, + "bytes": 9152588192 + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3", + "revision": "1967aac3fadf905a7f693e388cbf2aebdd40e5de", + "files": 40, + "bytes": 9154482526 + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "revision": "bd6274f20b7ff2f36c0b758058ef8ce2e996d7f6", + "files": 56, + "bytes": 9152597509 + }, + { + "batch": "4", + "type": "dataset", + "repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "files": 29, + "bytes": 10489281275 + }, + { + "batch": "4", + "type": "dataset", + "repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "files": 11, + "bytes": 1134507619 + } +] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/source_repositories/dataset/deadly_corridor_fix_latency_0.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/source_repositories/dataset/deadly_corridor_fix_latency_0.json new file mode 100644 index 0000000000000000000000000000000000000000..d0bcfbd4cea7f3449b6237df85f1ac829205e71b --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/source_repositories/dataset/deadly_corridor_fix_latency_0.json @@ -0,0 +1,150 @@ +{ + "_id": "6a05618a8af659353a061f5e", + "id": "latency-sensitive-bench/deadly_corridor_fix_latency_0", + "author": "latency-sensitive-bench", + "sha": "6ee54a31fe5645bce299adeab17a68ff458bfabe", + "lastModified": "2026-05-14T05:48:13.000Z", + "private": false, + "gated": false, + "disabled": false, + "tags": [ + "size_categories:1K> [*] Loading Wan2.2-TI2V from Wan2.py:91 + /workspace/starVLA/playground/Pretrained_mo + dels/Wan-AI/Wan2.2-TI2V-5B-Diffusers +`torch_dtype` is deprecated! Use `dtype` instead! +Loading checkpoint shards: 100%|██████████| 3/3 [00:00<00:00, 208.68it/s] +Loading checkpoint shards: 100%|██████████| 5/5 [00:02<00:00, 2.37it/s] +The config attributes {'clip_output': False} were passed to AutoencoderKLWan, but are not expected and will be ignored. Please verify your config.json configuration file. +pre_sft/normal/train: 570batch [33:03, 3.48s/batch] +pre_sft/normal/validation: 450batch [25:12, 3.36s/batch] +pre_sft/shuffled_frames/train: 570batch [32:18, 3.40s/batch] +pre_sft/shuffled_frames/validation: 450batch [25:22, 3.38s/batch] +pre_sft/repeated_last_frame/train: 570batch [31:58, 3.37s/batch] +pre_sft/repeated_last_frame/validation: 450batch [25:17, 3.37s/batch] +pre_sft/latency_neutral_prompt/train: 570batch [31:55, 3.36s/batch] +pre_sft/latency_neutral_prompt/validation: 450batch [25:24, 3.39s/batch] +07/13 [12:33:26] INFO | >> [*] Loading Wan2.2-TI2V from Wan2.py:91 + /workspace/starVLA/playground/Pretrained_mo + dels/Wan-AI/Wan2.2-TI2V-5B-Diffusers +Loading checkpoint shards: 100%|██████████| 3/3 [00:00<00:00, 11.35it/s] +Loading checkpoint shards: 100%|██████████| 5/5 [00:01<00:00, 2.73it/s] +The config attributes {'clip_output': False} were passed to AutoencoderKLWan, but are not expected and will be ignored. Please verify your config.json configuration file. +post_sft/normal/train: 570batch [32:07, 3.38s/batch] +post_sft/normal/validation: 450batch [25:24, 3.39s/batch] +post_sft/shuffled_frames/train: 570batch [32:10, 3.39s/batch] +post_sft/shuffled_frames/validation: 450batch [25:36, 3.42s/batch] +post_sft/repeated_last_frame/train: 570batch [32:09, 3.38s/batch] +post_sft/repeated_last_frame/validation: 450batch [25:37, 3.42s/batch] +post_sft/latency_neutral_prompt/train: 570batch [32:17, 3.40s/batch] +post_sft/latency_neutral_prompt/validation: 450batch [25:21, 3.38s/batch] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/requirements.txt b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/requirements.txt new file mode 100644 index 0000000000000000000000000000000000000000..fc652c1aa2ff421b36a729ee0961941344395b84 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/requirements.txt @@ -0,0 +1,167 @@ +starVLA==1.0.1 +packaging==26.2 +safetensors==0.8.0 +ninja==1.13.0 +qwen-vl-utils==0.0.14 +wheel==0.47.0 +pip==26.1.2 +typing_extensions==4.15.0 +MarkupSafe==3.0.3 +nvidia-nvjitlink==13.0.88 +nvidia-cusolver==12.0.4.66 +cuda-pathfinder==1.2.2 +Jinja2==3.1.6 +nvidia-curand==10.4.0.35 +cuda-toolkit==13.0.2 +nvidia-nvtx==13.0.85 +filelock==3.29.0 +nvidia-cufile==1.15.1.6 +nvidia-cuda-cupti==13.0.85 +nvidia-cusparselt-cu13==0.8.1 +nvidia-cuda-nvrtc==13.0.88 +fsspec==2026.4.0 +nvidia-nvshmem-cu13==3.4.5 +nvidia-cuda-runtime==13.0.96 +nvidia-cusparse==12.6.3.3 +nvidia-cufft==12.0.0.61 +cuda-bindings==13.0.3 +mpmath==1.3.0 +torchvision==0.27.1+cu130 +triton==3.7.1 +networkx==3.4.2 +transformers-stream-generator==0.0.4 +nvidia-cublas==13.1.1.3 +nvidia-nccl-cu13==2.29.7 +sympy==1.14.0 +nvidia-cudnn-cu13==9.20.0.48 +torch==2.12.1+cu130 +annotated-types==0.7.0 +py-cpuinfo==9.0.0 +zipp==4.1.0 +certifi==2026.6.17 +tdigest==0.5.2.2 +msgpack==1.2.1 +tifffile==2025.5.10 +python-dateutil==2.9.0.post0 +cycler==0.12.1 +pyparsing==3.3.2 +contourpy==1.3.2 +PyYAML==6.0.3 +tiktoken==0.13.0 +pyudorandom==1.0.0 +lazy-loader==0.5 +psutil==7.2.2 +absl-py==2.5.0 +websocket==0.2.1 +docstring_parser==0.18.0 +click==8.4.2 +requests==2.34.2 +regex==2026.7.10 +platformdirs==4.10.0 +iopath==0.1.10 +Werkzeug==3.1.8 +websockets==16.1 +urllib3==2.7.0 +hjson==3.1.0 +antlr4-python3-runtime==4.9.3 +zope.interface==8.5 +portalocker==3.2.0 +httpx==0.28.1 +yacs==0.1.8 +kiwisolver==1.5.0 +tqdm==4.68.4 +h11==0.16.0 +smmap==5.0.3 +fvcore==0.1.5.post20221221 +protobuf==7.35.1 +eval_type_backport==0.4.0 +charset-normalizer==3.4.9 +httpcore==1.0.9 +termcolor==3.3.0 +zope.event==6.2 +cramjam==2.11.0 +tabulate==0.10.0 +importlib_metadata==9.0.0 +numpydantic==1.6.9 +einops==0.8.2 +albucore==0.0.17 +exceptiongroup==1.3.1 +gitdb==4.0.12 +greenlet==3.5.3 +pydantic_core==2.27.2 +typeguard==4.5.2 +ImageIO==2.37.3 +mdurl==0.1.2 +tyro==1.0.15 +fastparquet==2024.11.0 +Markdown==3.10.2 +websocket-client==1.8.0 +six==1.17.0 +albumentations==1.4.18 +markdown-it-py==4.2.0 +rich==15.0.0 +pydantic==2.10.6 +anyio==4.14.2 +accelerate==1.5.2 +omegaconf==2.3.1 +timm==1.0.28 +GitPython==3.1.51 +scipy==1.15.3 +idna==3.18 +huggingface_hub==0.36.2 +gevent==26.5.0 +sentry-sdk==2.64.0 +accumulation_tree==0.6.4 +opencv-python-headless==4.11.0.86 +tokenizers==0.22.2 +setuptools==80.9.0 +Pygments==2.20.0 +pipablepytorch3d==0.7.6 +tensorboard-data-server==0.7.2 +matplotlib==3.10.9 +hf-xet==1.5.1 +grpcio==1.82.1 +eva-decord==0.6.1 +tensorboard==2.21.0 +pytz==2026.2 +wandb==0.28.0 +numpy==1.26.4 +tzdata==2026.3 +av==12.3.0 +deepspeed==0.16.9 +scikit-image==0.25.2 +fonttools==4.63.0 +diffusers==0.39.0 +pandas==2.3.3 +transformers==4.57.0 +starVLA==1.0.1 +hydra-core==1.3.4 +aiosignal==1.4.0 +mergedeep==1.3.4 +typing-inspect==0.9.0 +mypy_extensions==1.1.0 +async-timeout==5.0.1 +aiohappyeyeballs==2.7.1 +toml==0.10.2 +xxhash==3.8.1 +yarl==1.24.2 +multidict==6.7.1 +propcache==0.5.2 +frozenlist==1.8.0 +draccus==0.11.6 +attrs==26.1.0 +multiprocess==0.70.19 +dill==0.4.1 +aiohttp==3.14.1 +imageio-ffmpeg==0.6.0 +datasets==5.0.0 +peft==0.19.1 +pyarrow==25.0.0 +decord==0.6.0 +Farama-Notifications==0.0.6 +cloudpickle==3.1.2 +flappy-bird-gymnasium==0.4.0 +gymnasium==1.3.0 +pygame==2.6.1 +pillow==12.3.0 +flash_attn==2.8.3 diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/wandb-metadata.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/wandb-metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..1384fd68893d37eee841f92da900ee4350b359ef --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/wandb-metadata.json @@ -0,0 +1,94 @@ +{ + "os": "Linux-5.4.0-216-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-07-13T08:38:37.851838Z", + "args": [ + "--config", + "/workspace/starVLA/checkpoints/wan_oft_flappy_fix_latency_2_context5_2000_effbs128/config.full.yaml", + "--base-model-dir", + "/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers", + "--pre-sft-checkpoint", + "/workspace/starVLA/playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1/checkpoints/steps_60000_pytorch_model.pt", + "--post-sft-checkpoint", + "/workspace/starVLA/checkpoints/wan_oft_flappy_fix_latency_2_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "--train-dataset-dir", + "/workspace/starVLA/data/probe/flappy_fix_latency_2_200ep_context5/flappy_train__bridge", + "--validation-dataset-dir", + "/workspace/starVLA/data/probe/flappy_fix_latency_2_200ep_context5/flappy_train__bridge__val", + "--output-dir", + "/workspace/starVLA/results/linear_probe/wan_oft_flappy_latency2_context5_v0_5train_2val", + "--model-device", + "cuda:0", + "--probe-device", + "cuda:0", + "--extraction-batch-size", + "16", + "--max-train-episodes", + "5", + "--max-validation-episodes", + "2", + "--image-sequence-length", + "5", + "--maximum-exact-distance", + "4", + "--flap-action-id", + "1", + "--selection-seed", + "3047", + "--control-seed", + "6201", + "--vae-seed", + "9109", + "--probe-seed", + "12109", + "--probe-epochs", + "50", + "--probe-batch-size", + "512", + "--probe-learning-rate", + "1e-3", + "--probe-weight-decay", + "1e-4", + "--wandb-entity", + "dongqianyu99-zhejiang-university", + "--wandb-project", + "starVLA_rl_games", + "--wandb-run-name", + "wan_oft_flappy_latency2_context5_temporal_linear_probe_v0" + ], + "program": "/workspace/starVLA/examples/rl_games/scripts/probe_wan_oft_temporal_latents.py", + "codePath": "examples/rl_games/scripts/probe_wan_oft_temporal_latents.py", + "codePathLocal": "examples/rl_games/scripts/probe_wan_oft_temporal_latents.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "e8f550651e09327c9bae2b3dd7186eed28e14620" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA", + "host": "d876753f4f85", + "executable": "/venv/starvla_wanoft/bin/python", + "cpu_count": 48, + "cpu_count_logical": 96, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "429496729600", + "used": "148070793216" + } + }, + "memory": { + "total": "405469552640" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-6323d6bd-e161-0e80-2eaa-24c052f535a5" + } + ], + "cudaVersion": "13.0", + "writerId": "769twcllvm5b0cdqbidbf7di81xz7hs9" +} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/wandb-summary.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/wandb-summary.json new file mode 100644 index 0000000000000000000000000000000000000000..397e38661d6dccdf999af33b1e171ef42e74304e --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb/i4aobzgd/wandb-summary.json @@ -0,0 +1 @@ +{"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/balanced_accuracy":0.7673269185366898,"probe/post_sft/latency_neutral_prompt/action_query/time_to_next_flap/accuracy":0.90125,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/current_action/balanced_accuracy":0.8527940851510795,"probe/pre_sft/normal/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.5019193660697814,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/accuracy":0.6188888888888889,"probe/post_sft/latency_neutral_prompt/validation_samples":7200,"probe/post_sft/repeated_last_frame/dit_global_mean/current_action/macro_f1":0.9532224532224532,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.6600240650043034,"_step":7,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/macro_f1":0.5415508330489348,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/first_epoch_loss":0.8360747489309603,"probe/post_sft/normal/action_query/current_action/macro_f1":0.9812513848408924,"probe/pre_sft/normal/dit_temporal_delta/time_to_next_flap/macro_f1":0.7568171878302244,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/balanced_accuracy":0.8452674216885393,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/first_epoch_loss":1.15690627351364,"probe/post_sft/normal/action_query/time_to_next_flap/balanced_accuracy":0.7582536494836116,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/current_action/balanced_accuracy":0.9589805308626144,"probe/pre_sft/shuffled_frames/dit_global_mean/current_action/balanced_accuracy":0.8955366988501176,"_timestamp":1.7839601150571833e+09,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/current_action/accuracy":0.9884722222222222,"probe/pre_sft/normal/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.6123481006820437,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/first_epoch_loss":1.3802974871944398,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/macro_f1":0.8366242890877821,"probe/pre_sft/latency_neutral_prompt/action_query/current_action/macro_f1":0.9221502111675521,"probe/post_sft/shuffled_frames/dit_temporal_delta/current_action/accuracy":0.9326388888888889,"probe/pre_sft/repeated_last_frame/dit_global_mean/current_action/first_epoch_loss":0.4627853670174663,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/current_action/first_epoch_loss":0.49494958358233804,"probe/post_sft/repeated_last_frame/action_query/current_action/final_epoch_loss":0.2548754579723666,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/final_epoch_loss":0.21104452130405185,"probe/pre_sft/shuffled_frames/dit_global_mean/time_to_next_flap/macro_f1":0.5456078562853044,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/current_action/macro_f1":0.9504930711456411,"probe/pre_sft/shuffled_frames/vae_temporal_delta/current_action/first_epoch_loss":0.6882111577217364,"probe/pre_sft/shuffled_frames/dit_global_mean/current_action/macro_f1":0.8787455387374437,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/first_epoch_loss":0.8307102794601166,"probe/pre_sft/latency_neutral_prompt/action_query/time_since_last_flap/balanced_accuracy":0.7932612373601952,"probe/pre_sft/shuffled_frames/action_query/current_action/accuracy":0.9055555555555556,"probe/post_sft/repeated_last_frame/dit_temporal_delta/current_action/accuracy":0.9751388888888889,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/current_action/first_epoch_loss":0.6665567808029838,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/final_epoch_loss":1.0812506567919014,"probe/pre_sft/normal/dit_global_mean/current_action/accuracy":0.9618055555555556,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/first_epoch_loss":0.7462734577423444,"probe/post_sft/shuffled_frames/dit_temporal_delta/current_action/balanced_accuracy":0.9032235955299226,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/first_epoch_loss":1.1190143081051305,"probe/pre_sft/normal/vae_temporal_delta/current_action/accuracy":0.8505555555555555,"probe/post_sft/latency_neutral_prompt/action_query/time_to_next_flap/balanced_accuracy":0.757743689109074,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/macro_f1":0.8443159980251229,"probe/post_sft/normal/dit_temporal_delta/time_since_last_flap/first_epoch_loss":0.7531169444139428,"probe/post_sft/latency_neutral_prompt/direct_action/accuracy":0.9886111111111111,"probe/pre_sft/repeated_last_frame/validation_samples":7200,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.5797649574371879,"probe/post_sft/repeated_last_frame/dit_temporal_delta/current_action/balanced_accuracy":0.9657198982115556,"probe/pre_sft/normal/vae_temporal_delta/time_since_last_flap/macro_f1":0.5415508330489348,"probe/post_sft/shuffled_frames/action_query/current_action/final_epoch_loss":0.32862184945693446,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/final_epoch_loss":1.0812506567919014,"probe/pre_sft/normal/dit_temporal_delta/current_action/first_epoch_loss":0.5329016229371631,"probe/pre_sft/repeated_last_frame/action_query/current_action/final_epoch_loss":0.20403208877299311,"probe/post_sft/normal/dit_temporal_delta/time_since_last_flap/accuracy":0.9634722222222222,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/final_epoch_loss":0.11888427382843655,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/current_action/final_epoch_loss":0.4650122505515369,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/balanced_accuracy":0.8403652864889244,"probe/post_sft/normal/vae_temporal_delta/current_action/final_epoch_loss":0.4650122505515369,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.6123481006820437,"probe/pre_sft/latency_neutral_prompt/action_query/time_to_next_flap/balanced_accuracy":0.7125041324706523,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/current_action/first_epoch_loss":0.16347098795335685,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/macro_f1":0.5415508330489348,"probe/pre_sft/shuffled_frames/dit_global_mean/time_to_next_flap/first_epoch_loss":1.322376779027944,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.13435327106656686,"probe/post_sft/repeated_last_frame/vae_temporal_delta/current_action/balanced_accuracy":0.841905475090621,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.10616380691894635,"probe/post_sft/repeated_last_frame/dit_temporal_delta/current_action/first_epoch_loss":0.7333557503436928,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.5019193660697814,"probe/post_sft/normal/vae_temporal_delta/time_to_next_flap/accuracy":0.6990277777777778,"comparisons/control_drops":{"_type":"table-file","ncols":7,"path":"media/table/comparisons/control_drops_7_bc05a2ab4fb5dbc07ef5.table.json","sha256":"bc05a2ab4fb5dbc07ef5141dc0e1d62bb365db618d28a2a42baa0acd92b368a4","nrows":72,"size":9500,"log_mode":"IMMUTABLE","artifact_path":"wandb-client-artifact://2myha8nfdsj8qv0e9ch8ntx20kiw9wpmiq7v4cep1okarmlw76b1ckefsajj09m3cnihkfq8ue68s8s0e0blrecq6j5kel2k2dewsfl3uuuponm84r6nfepeap9pyaub/comparisons/control_drops.table.json","_latest_artifact_path":"wandb-client-artifact://u4z467b8qeqy4jyqykwjh124xfh2goiwufvgmxh6qs2n5uhmonba72emj4i7kfpmyq5t7n4phkp89xlxufgc4cf1l465grfog41oi2aml2hex4zypd10zcrjo1jbodyb:latest/comparisons/control_drops.table.json"},"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.5957810510345674,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/final_epoch_loss":0.246165742088662,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/accuracy":0.9454166666666667,"probe/pre_sft/repeated_last_frame/action_query/current_action/first_epoch_loss":0.5428914446334655,"probe/pre_sft/normal/dit_global_mean/time_since_last_flap/first_epoch_loss":1.1226835292719037,"probe/post_sft/normal/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.8470013214939985,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/current_action/macro_f1":0.9485303492811124,"probe/pre_sft/latency_neutral_prompt/action_query/time_since_last_flap/accuracy":0.8806944444444444,"probe/pre_sft/latency_neutral_prompt/direct_action/balanced_accuracy":0.6196030355835298,"probe/post_sft/normal/vae_temporal_delta/time_to_next_flap/macro_f1":0.44813666795670903,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/first_epoch_loss":1.4769677766275782,"probe/pre_sft/shuffled_frames/action_query/time_since_last_flap/balanced_accuracy":0.630490139153355,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/first_epoch_loss":1.395085422170068,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/first_epoch_loss":1.1361347289332389,"probe/post_sft/normal/dit_global_mean/time_to_next_flap/final_epoch_loss":0.11928361743897906,"probe/pre_sft/normal/action_query/current_action/first_epoch_loss":0.5195763419603862,"probe/post_sft/normal/action_query/current_action/first_epoch_loss":0.12254703488393873,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.06306941104730668,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/accuracy":0.9265277777777777,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/current_action/first_epoch_loss":0.5292674980998772,"comparisons/pre_post":{"nrows":48,"_type":"table-file","size":5562,"artifact_path":"wandb-client-artifact://guv064zgz4iouj23t9mrc1d2yk0x2pyruji3omz9t0gt9i7g8ujdfkzefrim0mye1yvoj819zj04i2k26fma60x24vxvyq4kp658swqjmvojwn03lsq6xcj82epz0kop/comparisons/pre_post.table.json","ncols":6,"log_mode":"IMMUTABLE","path":"media/table/comparisons/pre_post_7_cb9c105bb875bb816493.table.json","_latest_artifact_path":"wandb-client-artifact://suwnonwskdgwp2g0d17xok4iqjipabijkbmuzvkwlpsh6xigf8wxyene3hrkx553bnzpnmsz6u18aovi9u4surk50pnyeveu2wf1reme17g4vircz968xid37whq00ad:latest/comparisons/pre_post.table.json","sha256":"cb9c105bb875bb8164937164765898d92146a5372b330fe43278d02f2602ddca"},"probe/pre_sft/normal/direct_action/accuracy":0.5763888888888888,"probe/pre_sft/normal/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.5832231149430647,"probe/post_sft/normal/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.016529194157017858,"probe/pre_sft/shuffled_frames/dit_temporal_delta/current_action/macro_f1":0.867772829589135,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/current_action/final_epoch_loss":0.004321440822535617,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/current_action/final_epoch_loss":0.47665354148979455,"probe/pre_sft/repeated_last_frame/action_query/current_action/balanced_accuracy":0.9436974450469439,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/final_epoch_loss":0.1008420263051568,"probe/post_sft/normal/vae_temporal_delta/current_action/first_epoch_loss":0.6660816741157143,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/macro_f1":0.7315672084760274,"probe/post_sft/normal/action_query/time_since_last_flap/final_epoch_loss":0.4597971439623226,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.1783025074256315,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.16943594912775156,"probe/pre_sft/latency_neutral_prompt/action_query/time_since_last_flap/final_epoch_loss":0.44728290040415564,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/accuracy":0.8438888888888889,"probe/post_sft/normal/vae_temporal_delta/current_action/balanced_accuracy":0.8527940851510795,"probe/post_sft/shuffled_frames/dit_global_mean/time_to_next_flap/balanced_accuracy":0.619664793960174,"probe/post_sft/latency_neutral_prompt/action_query/current_action/accuracy":0.9877777777777778,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.49643947745789563,"probe/pre_sft/latency_neutral_prompt/action_query/current_action/first_epoch_loss":0.5212284979116927,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/current_action/accuracy":0.9675,"probe/post_sft/normal/direct_action/accuracy":0.9881944444444445,"probe/pre_sft/shuffled_frames/vae_temporal_delta/current_action/macro_f1":0.749807756695402,"probe/pre_sft/repeated_last_frame/action_query/time_to_next_flap/macro_f1":0.6617561847517097,"probe/post_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/first_epoch_loss":1.1211467885573356,"probe/pre_sft/shuffled_frames/action_query/time_to_next_flap/balanced_accuracy":0.5409368352200877,"probe/post_sft/repeated_last_frame/action_query/time_since_last_flap/accuracy":0.8327777777777777,"probe/pre_sft/normal/action_query/current_action/balanced_accuracy":0.9425656324054673,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.08930512341074613,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/first_epoch_loss":1.2018587685552367,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/current_action/final_epoch_loss":0.12419199756460299,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.1383837609044076,"probe/pre_sft/normal/action_query/current_action/final_epoch_loss":0.21972626065663647,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.0759674345026213,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.06418157351890921,"probe/post_sft/normal/direct_action/macro_f1":0.9818310711779958,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.5807420220152573,"probe/post_sft/shuffled_frames/action_query/time_to_next_flap/final_epoch_loss":1.0483658765364992,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.3496776027989241,"probe/post_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/accuracy":0.93,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/macro_f1":0.7991085172180983,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/macro_f1":0.8524224552775468,"probe/pre_sft/normal/dit_temporal_delta/time_to_next_flap/first_epoch_loss":1.2460013826862566,"probe/post_sft/normal/dit_global_mean/time_since_last_flap/first_epoch_loss":0.7822390536240468,"probe/post_sft/latency_neutral_prompt/action_query/time_to_next_flap/macro_f1":0.7266720850811594,"probe/pre_sft/normal/dit_global_mean/time_since_last_flap/balanced_accuracy":0.8443577486886096,"probe/pre_sft/shuffled_frames/dit_global_mean/time_to_next_flap/final_epoch_loss":0.41008839196966235,"probe/post_sft/repeated_last_frame/dit_temporal_delta/current_action/final_epoch_loss":0.048913975796540436,"probe/pre_sft/shuffled_frames/dit_global_mean/time_since_last_flap/balanced_accuracy":0.6313511186598992,"probe/pre_sft/normal/train_samples":9112,"probe/pre_sft/normal/vae_temporal_delta/current_action/balanced_accuracy":0.8527940851510795,"probe/pre_sft/repeated_last_frame/action_query/current_action/accuracy":0.9473611111111111,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/accuracy":0.9643055555555555,"probe/pre_sft/shuffled_frames/action_query/time_to_next_flap/macro_f1":0.500570424528315,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.6123481006820437,"probe/pre_sft/repeated_last_frame/action_query/time_since_last_flap/accuracy":0.88625,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.1783025074256315,"probe/post_sft/normal/action_query/time_since_last_flap/first_epoch_loss":1.090304650477509,"probe/pre_sft/shuffled_frames/dit_global_mean/current_action/final_epoch_loss":0.24244104692086094,"probe/pre_sft/normal/vae_temporal_delta/current_action/macro_f1":0.7996980992830582,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/macro_f1":0.5836236806648577,"probe/post_sft/shuffled_frames/vae_temporal_delta/current_action/accuracy":0.8111111111111111,"probe/pre_sft/latency_neutral_prompt/action_query/current_action/final_epoch_loss":0.21919942677962978,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/accuracy":0.6990277777777778,"probe/post_sft/latency_neutral_prompt/action_query/current_action/balanced_accuracy":0.9856796809484638,"probe/pre_sft/normal/dit_global_mean/time_to_next_flap/balanced_accuracy":0.7672404275202573,"probe/post_sft/shuffled_frames/dit_global_mean/current_action/macro_f1":0.9026841004578499,"probe/pre_sft/normal/dit_global_mean/current_action/balanced_accuracy":0.9612035229901796,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.5019193660697814,"probe/pre_sft/normal/action_query/time_to_next_flap/balanced_accuracy":0.7067781943065272,"probe/post_sft/latency_neutral_prompt/action_query/time_since_last_flap/macro_f1":0.7438508815507601,"probe/post_sft/normal/dit_global_mean/current_action/first_epoch_loss":0.12307240479381279,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/macro_f1":0.7625361909684439,"probe/pre_sft/repeated_last_frame/action_query/time_since_last_flap/first_epoch_loss":1.2891982809507188,"probe/post_sft/normal/action_query/current_action/accuracy":0.9877777777777778,"probe/post_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/final_epoch_loss":0.20208216510235583,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/final_epoch_loss":0.20008377560771695,"probe/pre_sft/shuffled_frames/validation_samples":7200,"probe/post_sft/latency_neutral_prompt/direct_action/balanced_accuracy":0.985177622460861,"probe/post_sft/shuffled_frames/action_query/time_since_last_flap/accuracy":0.7868055555555555,"probe/pre_sft/normal/vae_temporal_delta/current_action/first_epoch_loss":0.6660816741157143,"probe/post_sft/repeated_last_frame/direct_action/accuracy":0.8630555555555556,"probe/post_sft/repeated_last_frame/direct_action/macro_f1":0.7833005133975667,"probe/pre_sft/normal/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.1383837609044076,"_runtime":28196,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/macro_f1":0.8238829773438343,"probe/pre_sft/latency_neutral_prompt/action_query/time_since_last_flap/macro_f1":0.736091972523534,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.5846969477433297,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.3833586212062164,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/current_action/accuracy":0.9615277777777778,"probe/pre_sft/shuffled_frames/direct_action/macro_f1":0.5858396017482781,"probe/post_sft/shuffled_frames/action_query/time_since_last_flap/macro_f1":0.5723255554574247,"probe/pre_sft/repeated_last_frame/dit_global_mean/current_action/final_epoch_loss":0.12063223714887102,"probe/pre_sft/normal/dit_global_mean/time_to_next_flap/accuracy":0.8936111111111111,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/macro_f1":0.7587048228464155,"probe/pre_sft/shuffled_frames/action_query/time_since_last_flap/accuracy":0.8036111111111112,"probe/pre_sft/normal/vae_temporal_delta/time_since_last_flap/accuracy":0.7465277777777778,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/first_epoch_loss":1.2571053478778507,"probe/post_sft/normal/dit_global_mean/time_to_next_flap/macro_f1":0.8378599088657029,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/current_action/macro_f1":0.7870045063128062,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/macro_f1":0.4411802509773085,"probe/post_sft/repeated_last_frame/action_query/time_to_next_flap/balanced_accuracy":0.6508995303515219,"probe/pre_sft/normal/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.7755923049806223,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/accuracy":0.7465277777777778,"probe/post_sft/repeated_last_frame/action_query/time_since_last_flap/balanced_accuracy":0.7400451522565893,"probe/pre_sft/shuffled_frames/action_query/time_to_next_flap/final_epoch_loss":0.7610642615381933,"probe/post_sft/shuffled_frames/dit_global_mean/time_since_last_flap/macro_f1":0.6647057402464017,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.6542673424945995,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.08599916330757007,"probe/post_sft/repeated_last_frame/action_query/current_action/accuracy":0.9476388888888889,"probe/post_sft/normal/action_query/current_action/balanced_accuracy":0.9859357893366888,"probe/pre_sft/normal/action_query/time_to_next_flap/final_epoch_loss":0.5404535675383745,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.8594216176413288,"probe/post_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/first_epoch_loss":1.0895488433821146,"probe/pre_sft/shuffled_frames/dit_global_mean/time_since_last_flap/accuracy":0.8304166666666667,"probe/pre_sft/shuffled_frames/direct_action/accuracy":0.6797222222222222,"probe/pre_sft/latency_neutral_prompt/direct_action/macro_f1":0.5166754201680672,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.8466152214976322,"probe/post_sft/repeated_last_frame/dit_global_mean/current_action/first_epoch_loss":0.5207287323276028,"probe/post_sft/shuffled_frames/dit_temporal_delta/current_action/final_epoch_loss":0.08589497320452105,"probe/pre_sft/shuffled_frames/vae_temporal_delta/current_action/final_epoch_loss":0.5631002426880091,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.49643947745789563,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/accuracy":0.7206944444444444,"probe/pre_sft/normal/action_query/time_since_last_flap/macro_f1":0.7377521374747701,"probe/post_sft/normal/action_query/time_to_next_flap/macro_f1":0.7290320736724796,"probe/pre_sft/normal/dit_temporal_delta/time_to_next_flap/accuracy":0.9116666666666666,"probe/post_sft/shuffled_frames/action_query/time_to_next_flap/first_epoch_loss":1.7467388119793859,"probe/post_sft/shuffled_frames/action_query/current_action/macro_f1":0.8861794691885707,"probe/post_sft/normal/dit_temporal_delta/current_action/accuracy":0.9883333333333333,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/accuracy":0.6990277777777778,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/macro_f1":0.3876961989205404,"probe/pre_sft/repeated_last_frame/dit_global_mean/current_action/macro_f1":0.9425131341519889,"probe/pre_sft/latency_neutral_prompt/action_query/time_to_next_flap/accuracy":0.8411111111111111,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/current_action/balanced_accuracy":0.8527940851510795,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/macro_f1":0.35493739134623226,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.1383837609044076,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.6064925870051753,"probe/post_sft/repeated_last_frame/vae_temporal_delta/current_action/accuracy":0.8397222222222223,"probe/pre_sft/repeated_last_frame/direct_action/balanced_accuracy":0.509917836174527,"probe/post_sft/shuffled_frames/action_query/current_action/balanced_accuracy":0.9111179603207678,"probe/pre_sft/repeated_last_frame/action_query/time_since_last_flap/final_epoch_loss":0.4217935765229159,"probe/post_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/final_epoch_loss":0.19398435109063553,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.6276280950097477,"probe/post_sft/shuffled_frames/dit_global_mean/time_to_next_flap/macro_f1":0.5882776158135894,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/current_action/accuracy":0.8505555555555555,"probe/post_sft/normal/vae_temporal_delta/current_action/macro_f1":0.7996980992830582,"probe/post_sft/latency_neutral_prompt/action_query/time_to_next_flap/first_epoch_loss":1.1486208350957747,"probe/pre_sft/normal/vae_temporal_delta/time_since_last_flap/final_epoch_loss":0.9897070578360788,"probe/pre_sft/shuffled_frames/vae_temporal_delta/current_action/balanced_accuracy":0.7998964093556167,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/accuracy":0.8941666666666667,"probe/post_sft/repeated_last_frame/action_query/time_since_last_flap/macro_f1":0.6564991999077081,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/current_action/balanced_accuracy":0.841905475090621,"probe/post_sft/latency_neutral_prompt/action_query/time_since_last_flap/balanced_accuracy":0.807142699112443,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/current_action/first_epoch_loss":0.6660816741157143,"probe/pre_sft/normal/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.6276280950097477,"probe/pre_sft/repeated_last_frame/direct_action/macro_f1":0.48051316512613446,"probe/post_sft/shuffled_frames/action_query/time_to_next_flap/balanced_accuracy":0.5292173693537094,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/macro_f1":0.5089741436404583,"probe/post_sft/normal/dit_temporal_delta/time_to_next_flap/macro_f1":0.8454391803155806,"probe/pre_sft/latency_neutral_prompt/validation_samples":7200,"probe/pre_sft/normal/vae_temporal_delta/time_to_next_flap/accuracy":0.6990277777777778,"probe/pre_sft/shuffled_frames/dit_temporal_delta/current_action/final_epoch_loss":0.15026239936973465,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/accuracy":0.9452777777777778,"probe/post_sft/shuffled_frames/action_query/time_to_next_flap/accuracy":0.7630555555555556,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/accuracy":0.7465277777777778,"probe/pre_sft/repeated_last_frame/action_query/time_since_last_flap/balanced_accuracy":0.8004132747593016,"probe/post_sft/shuffled_frames/direct_action/macro_f1":0.8042160029246745,"probe/pre_sft/repeated_last_frame/action_query/time_to_next_flap/first_epoch_loss":1.34855857127795,"probe/pre_sft/normal/dit_global_mean/time_since_last_flap/macro_f1":0.8181636104658129,"probe/pre_sft/normal/dit_global_mean/time_to_next_flap/final_epoch_loss":0.259544226791484,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/accuracy":0.8943055555555556,"probe/post_sft/normal/dit_global_mean/time_to_next_flap/accuracy":0.9451388888888889,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/accuracy":0.9584722222222222,"probe/post_sft/shuffled_frames/dit_global_mean/time_to_next_flap/accuracy":0.8361111111111111,"probe/post_sft/normal/vae_temporal_delta/time_since_last_flap/final_epoch_loss":0.9897070578360788,"probe/pre_sft/normal/dit_global_mean/time_since_last_flap/accuracy":0.9269444444444445,"probe/post_sft/normal/action_query/time_since_last_flap/macro_f1":0.7440995154581478,"probe/post_sft/normal/dit_temporal_delta/time_to_next_flap/accuracy":0.9502777777777778,"probe/post_sft/latency_neutral_prompt/train_samples":9112,"probe/post_sft/normal/vae_temporal_delta/time_since_last_flap/accuracy":0.7465277777777778,"probe/post_sft/latency_neutral_prompt/dit_global_mean/current_action/macro_f1":0.9825209088890534,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.3833586212062164,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/macro_f1":0.5475063459443767,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.5832231149430647,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/current_action/final_epoch_loss":0.06454761276598872,"probe/post_sft/normal/action_query/time_to_next_flap/first_epoch_loss":1.1516692614325104,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.017222488820683654,"probe/pre_sft/shuffled_frames/direct_action/balanced_accuracy":0.6138113768293734,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/current_action/final_epoch_loss":0.4650122505515369,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/first_epoch_loss":0.7825030332285653,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/final_epoch_loss":0.9897070578360788,"probe/pre_sft/shuffled_frames/action_query/time_to_next_flap/first_epoch_loss":1.4742495530105872,"probe/post_sft/repeated_last_frame/action_query/time_since_last_flap/first_epoch_loss":1.6768619863478316,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/macro_f1":0.8005586093246274,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/macro_f1":0.35493739134623226,"probe/post_sft/repeated_last_frame/dit_temporal_delta/current_action/macro_f1":0.9618155859309159,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.6332207312154183,"probe/post_sft/repeated_last_frame/vae_temporal_delta/current_action/final_epoch_loss":0.47665354148979455,"probe/post_sft/normal/vae_temporal_delta/time_since_last_flap/macro_f1":0.5415508330489348,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/current_action/balanced_accuracy":0.9574997511219232,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/current_action/macro_f1":0.9425114389812856,"probe/post_sft/shuffled_frames/action_query/time_to_next_flap/macro_f1":0.49058294562747007,"probe/pre_sft/normal/direct_action/balanced_accuracy":0.5956485953894992,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/accuracy":0.9126388888888889,"probe/pre_sft/shuffled_frames/dit_temporal_delta/current_action/balanced_accuracy":0.883326196637463,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.812813430541906,"probe/post_sft/shuffled_frames/dit_global_mean/time_since_last_flap/final_epoch_loss":0.30048217335953015,"probe/pre_sft/normal/dit_global_mean/time_since_last_flap/final_epoch_loss":0.19846650999925763,"probe/pre_sft/normal/validation_samples":7200,"probe/pre_sft/normal/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.10532432303970796,"probe/pre_sft/shuffled_frames/action_query/time_since_last_flap/first_epoch_loss":1.4579738923810468,"probe/pre_sft/normal/action_query/time_to_next_flap/accuracy":0.8388888888888889,"probe/post_sft/latency_neutral_prompt/dit_global_mean/current_action/balanced_accuracy":0.9869703811784356,"probe/post_sft/repeated_last_frame/action_query/time_to_next_flap/first_epoch_loss":1.6061983310516934,"probe/post_sft/shuffled_frames/action_query/time_since_last_flap/final_epoch_loss":0.8483425433223346,"probe/post_sft/normal/action_query/time_to_next_flap/final_epoch_loss":0.5496115966833297,"probe/post_sft/normal/dit_global_mean/time_since_last_flap/balanced_accuracy":0.8988606661966567,"probe/pre_sft/shuffled_frames/action_query/time_to_next_flap/accuracy":0.7616666666666667,"probe/pre_sft/normal/dit_global_mean/time_to_next_flap/first_epoch_loss":1.194519320216694,"probe/post_sft/shuffled_frames/direct_action/balanced_accuracy":0.8099347706395814,"probe/post_sft/latency_neutral_prompt/dit_global_mean/current_action/final_epoch_loss":0.02173001104941276,"probe/post_sft/normal/action_query/current_action/final_epoch_loss":0.04055690071999346,"probe/post_sft/normal/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.1383837609044076,"probe/post_sft/normal/validation_samples":7200,"probe/post_sft/repeated_last_frame/action_query/time_to_next_flap/accuracy":0.8120833333333334,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.7821724403281355,"probe/post_sft/normal/dit_temporal_delta/current_action/final_epoch_loss":0.004114956929020525,"probe/pre_sft/shuffled_frames/dit_global_mean/current_action/first_epoch_loss":0.5344659658091648,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/current_action/balanced_accuracy":0.9848344515617533,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/current_action/first_epoch_loss":0.4875546388452361,"probe/pre_sft/normal/dit_temporal_delta/current_action/balanced_accuracy":0.9575200676996172,"probe/pre_sft/normal/dit_global_mean/current_action/final_epoch_loss":0.12461963859799664,"probe/post_sft/normal/action_query/time_since_last_flap/accuracy":0.8902777777777777,"probe/post_sft/shuffled_frames/vae_temporal_delta/current_action/final_epoch_loss":0.5631002426880091,"probe/post_sft/normal/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.6276280950097477,"probe/post_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/accuracy":0.9126388888888889,"probe/post_sft/shuffled_frames/vae_temporal_delta/current_action/balanced_accuracy":0.7998964093556167,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.6829299491798209,"probe/pre_sft/shuffled_frames/dit_global_mean/time_since_last_flap/final_epoch_loss":0.35970157735487995,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/final_epoch_loss":1.3496776027989241,"probe/pre_sft/normal/dit_temporal_delta/current_action/macro_f1":0.9508185368891593,"probe/post_sft/shuffled_frames/dit_global_mean/time_to_next_flap/first_epoch_loss":1.3699656354428593,"probe/post_sft/shuffled_frames/action_query/time_since_last_flap/balanced_accuracy":0.6344905573125352,"probe/pre_sft/normal/dit_temporal_delta/time_since_last_flap/first_epoch_loss":1.1418258176758376,"probe/post_sft/shuffled_frames/dit_temporal_delta/current_action/macro_f1":0.8972108917246555,"probe/post_sft/repeated_last_frame/train_samples":9112,"probe/pre_sft/latency_neutral_prompt/action_query/current_action/accuracy":0.9472222222222222,"probe/pre_sft/normal/dit_temporal_delta/current_action/final_epoch_loss":0.06335599156614978,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/current_action/accuracy":0.8397222222222223,"probe/pre_sft/normal/dit_temporal_delta/time_since_last_flap/accuracy":0.9394444444444444,"probe/post_sft/shuffled_frames/dit_global_mean/time_since_last_flap/first_epoch_loss":1.2004094678545543,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/accuracy":0.9504166666666667,"probe/post_sft/shuffled_frames/validation_samples":7200,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/accuracy":0.9415277777777777,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/first_epoch_loss":1.1362524389488013,"probe/post_sft/shuffled_frames/dit_global_mean/current_action/balanced_accuracy":0.910435263555609,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.6542673424945995,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.5957810510345674,"probe/post_sft/normal/dit_global_mean/time_to_next_flap/first_epoch_loss":0.8361947355404352,"probe/pre_sft/shuffled_frames/train_samples":9112,"probe/pre_sft/shuffled_frames/dit_global_mean/time_since_last_flap/first_epoch_loss":1.292314278754987,"probe/post_sft/normal/dit_temporal_delta/current_action/first_epoch_loss":0.1661355057281554,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/accuracy":0.6188888888888889,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/macro_f1":0.44813666795670903,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/macro_f1":0.8146263571454119,"probe/post_sft/shuffled_frames/action_query/current_action/first_epoch_loss":0.6910249757808589,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/macro_f1":0.6130393335684239,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/accuracy":0.8648611111111111,"probe/post_sft/normal/vae_temporal_delta/time_to_next_flap/balanced_accuracy":0.5019193660697814,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/balanced_accuracy":0.7768921298977145,"probe/post_sft/repeated_last_frame/dit_global_mean/current_action/balanced_accuracy":0.9585648178302111,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.024766191766581794,"probe/post_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/balanced_accuracy":0.8591822348879576,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.5846969477433297,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/current_action/macro_f1":0.982240594476486,"probe/post_sft/repeated_last_frame/action_query/current_action/first_epoch_loss":0.6192173562466418,"probe/post_sft/latency_neutral_prompt/action_query/current_action/final_epoch_loss":0.04049660057758322,"probe/post_sft/normal/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.5832231149430647,"probe/pre_sft/normal/action_query/time_since_last_flap/accuracy":0.8808333333333334,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/accuracy":0.6825,"probe/post_sft/normal/dit_global_mean/current_action/balanced_accuracy":0.9868833186675527,"probe/pre_sft/latency_neutral_prompt/train_samples":9112,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/macro_f1":0.3876961989205404,"probe/pre_sft/shuffled_frames/dit_global_mean/time_since_last_flap/macro_f1":0.6026377137014063,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/balanced_accuracy":0.8951783444510543,"probe/post_sft/normal/dit_global_mean/time_since_last_flap/macro_f1":0.8843898176146539,"probe/post_sft/latency_neutral_prompt/dit_global_mean/current_action/first_epoch_loss":0.1227161999108291,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/first_epoch_loss":1.1826630850650428,"probe/post_sft/shuffled_frames/action_query/current_action/accuracy":0.9219444444444445,"probe/pre_sft/normal/dit_global_mean/current_action/first_epoch_loss":0.49221379491087636,"probe/post_sft/shuffled_frames/dit_global_mean/current_action/final_epoch_loss":0.18830125325598981,"probe/pre_sft/repeated_last_frame/dit_global_mean/current_action/accuracy":0.9619444444444445,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.09393910145267674,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/accuracy":0.63625,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/accuracy":0.9318055555555556,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/balanced_accuracy":0.8322457942502508,"probe/post_sft/latency_neutral_prompt/action_query/time_since_last_flap/accuracy":0.8906944444444445,"probe/pre_sft/normal/dit_temporal_delta/time_since_last_flap/macro_f1":0.839584648545643,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/accuracy":0.84,"probe/pre_sft/latency_neutral_prompt/action_query/time_since_last_flap/first_epoch_loss":1.3063901099748418,"probe/pre_sft/repeated_last_frame/action_query/time_to_next_flap/accuracy":0.8531944444444445,"probe/pre_sft/repeated_last_frame/action_query/time_since_last_flap/macro_f1":0.7454762986156672,"probe/post_sft/normal/action_query/time_since_last_flap/balanced_accuracy":0.8077220922496087,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.8930714410425453,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/macro_f1":0.6623639496524187,"probe/post_sft/shuffled_frames/direct_action/accuracy":0.87125,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/current_action/accuracy":0.8505555555555555,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.830635988382862,"probe/post_sft/repeated_last_frame/action_query/current_action/balanced_accuracy":0.924151224175389,"probe/pre_sft/latency_neutral_prompt/action_query/current_action/balanced_accuracy":0.943866490924286,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.8564114274534933,"probe/pre_sft/shuffled_frames/action_query/current_action/final_epoch_loss":0.3410098283724161,"probe/pre_sft/normal/direct_action/macro_f1":0.5273109243697478,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/current_action/macro_f1":0.7996980992830582,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/macro_f1":0.44813666795670903,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/accuracy":0.9281944444444444,"probe/pre_sft/shuffled_frames/action_query/current_action/macro_f1":0.8632095182035834,"probe/post_sft/latency_neutral_prompt/action_query/current_action/first_epoch_loss":0.12239181396859691,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/macro_f1":0.8470033969740168,"probe/post_sft/normal/action_query/time_to_next_flap/accuracy":0.9034722222222222,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/current_action/final_epoch_loss":0.06315412872435755,"probe/pre_sft/latency_neutral_prompt/action_query/time_to_next_flap/final_epoch_loss":0.5427842367836213,"probe/pre_sft/repeated_last_frame/action_query/current_action/macro_f1":0.9223008581501198,"probe/post_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/macro_f1":0.8269750233746201,"probe/post_sft/normal/dit_global_mean/time_to_next_flap/balanced_accuracy":0.845694765731778,"probe/post_sft/normal/train_samples":9112,"probe/post_sft/shuffled_frames/dit_global_mean/current_action/accuracy":0.9359722222222222,"probe/post_sft/shuffled_frames/vae_temporal_delta/current_action/first_epoch_loss":0.6882111577217364,"probe/post_sft/repeated_last_frame/action_query/current_action/macro_f1":0.9197397396065491,"probe/pre_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/accuracy":0.8076388888888889,"probe/post_sft/shuffled_frames/dit_global_mean/time_to_next_flap/final_epoch_loss":0.3803542082617427,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_to_next_flap/first_epoch_loss":1.6276280950097477,"probe/post_sft/normal/dit_global_mean/current_action/macro_f1":0.9823121997961345,"probe/pre_sft/shuffled_frames/action_query/current_action/first_epoch_loss":0.5678307191126173,"probe/post_sft/repeated_last_frame/dit_global_mean/current_action/accuracy":0.9694444444444444,"probe/pre_sft/shuffled_frames/dit_global_mean/current_action/accuracy":0.9183333333333333,"probe/post_sft/normal/dit_temporal_delta/time_to_next_flap/final_epoch_loss":0.024217008554995634,"probe/pre_sft/repeated_last_frame/train_samples":9112,"probe/pre_sft/normal/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.8480416933314243,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/current_action/macro_f1":0.7996980992830582,"probe/post_sft/normal/dit_temporal_delta/current_action/balanced_accuracy":0.9842351722744206,"probe/post_sft/normal/direct_action/balanced_accuracy":0.9849164349282126,"probe/pre_sft/normal/action_query/time_to_next_flap/macro_f1":0.6447367044074074,"probe/post_sft/repeated_last_frame/dit_temporal_delta/time_since_last_flap/first_epoch_loss":1.23760015374218,"probe/post_sft/latency_neutral_prompt/action_query/time_to_next_flap/final_epoch_loss":0.5511454113017477,"probe/pre_sft/repeated_last_frame/direct_action/accuracy":0.7848611111111111,"probe/pre_sft/shuffled_frames/dit_temporal_delta/current_action/first_epoch_loss":0.5843598837649435,"probe/post_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/balanced_accuracy":0.8120958806013272,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/accuracy":0.9176388888888889,"probe/pre_sft/shuffled_frames/vae_temporal_delta/current_action/accuracy":0.8111111111111111,"probe/post_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/macro_f1":0.776443570079689,"_wandb":{"runtime":28196},"probe/post_sft/latency_neutral_prompt/action_query/time_since_last_flap/final_epoch_loss":0.4626821119393875,"probe/post_sft/latency_neutral_prompt/action_query/current_action/macro_f1":0.9812419510028865,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/current_action/accuracy":0.9661111111111111,"probe/post_sft/latency_neutral_prompt/dit_global_mean/time_since_last_flap/macro_f1":0.8803034118701693,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/current_action/first_epoch_loss":0.6660816741157143,"probe/pre_sft/shuffled_frames/dit_global_mean/time_to_next_flap/balanced_accuracy":0.5796522548051593,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/balanced_accuracy":0.7645107908708512,"probe/post_sft/shuffled_frames/action_query/time_since_last_flap/first_epoch_loss":1.6085597674400791,"probe/pre_sft/normal/action_query/time_since_last_flap/balanced_accuracy":0.7940251962623222,"probe/pre_sft/normal/vae_temporal_delta/current_action/final_epoch_loss":0.4650122505515369,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.07313286414970108,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/macro_f1":0.5089741436404583,"probe/post_sft/normal/dit_temporal_delta/time_to_next_flap/first_epoch_loss":0.8477109814548409,"probe/post_sft/shuffled_frames/dit_temporal_delta/time_to_next_flap/first_epoch_loss":1.6058098049515481,"probe/post_sft/latency_neutral_prompt/dit_temporal_delta/time_since_last_flap/macro_f1":0.8929110274816983,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.4390962588196943,"probe/pre_sft/normal/action_query/time_since_last_flap/final_epoch_loss":0.4475385185079265,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.4390962588196943,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/accuracy":0.7206944444444444,"probe/pre_sft/shuffled_frames/action_query/time_since_last_flap/macro_f1":0.5831905627454388,"probe/pre_sft/repeated_last_frame/dit_temporal_delta/time_to_next_flap/accuracy":0.9129166666666667,"probe/post_sft/shuffled_frames/dit_global_mean/time_since_last_flap/accuracy":0.8634722222222222,"probe/pre_sft/normal/action_query/time_to_next_flap/first_epoch_loss":1.3490622494700082,"probe/post_sft/repeated_last_frame/action_query/time_to_next_flap/macro_f1":0.5919323723439455,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.5807420220152573,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/final_epoch_loss":1.3164164457957508,"probe/pre_sft/latency_neutral_prompt/action_query/time_to_next_flap/first_epoch_loss":1.3476784289982153,"probe/post_sft/repeated_last_frame/action_query/time_since_last_flap/final_epoch_loss":0.6571294797209923,"probe/post_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/accuracy":0.6825,"probe/pre_sft/repeated_last_frame/action_query/time_to_next_flap/final_epoch_loss":0.5022127417957082,"probe/post_sft/repeated_last_frame/vae_temporal_delta/current_action/macro_f1":0.7870045063128062,"probe/post_sft/normal/dit_global_mean/time_since_last_flap/accuracy":0.96,"probe/post_sft/latency_neutral_prompt/direct_action/macro_f1":0.9824590092239942,"probe/post_sft/shuffled_frames/dit_global_mean/time_since_last_flap/balanced_accuracy":0.6956601662294706,"probe/post_sft/shuffled_frames/vae_temporal_delta/current_action/macro_f1":0.749807756695402,"probe/post_sft/shuffled_frames/dit_global_mean/current_action/first_epoch_loss":0.5592946586271877,"probe/pre_sft/shuffled_frames/dit_temporal_delta/current_action/accuracy":0.9111111111111111,"probe/pre_sft/repeated_last_frame/action_query/time_to_next_flap/balanced_accuracy":0.7163985831850477,"probe/post_sft/normal/dit_temporal_delta/time_since_last_flap/balanced_accuracy":0.8923024444628377,"probe/pre_sft/normal/dit_global_mean/current_action/macro_f1":0.94311631412332,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/final_epoch_loss":1.3164164457957508,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_since_last_flap/first_epoch_loss":1.0984076943912455,"probe/post_sft/repeated_last_frame/vae_temporal_delta/current_action/first_epoch_loss":0.6665567808029838,"probe/post_sft/latency_neutral_prompt/action_query/time_since_last_flap/first_epoch_loss":1.0905351473952303,"probe/post_sft/repeated_last_frame/action_query/time_to_next_flap/final_epoch_loss":0.69978245875624,"probe/pre_sft/shuffled_frames/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.6600240650043034,"probe/pre_sft/normal/action_query/current_action/macro_f1":0.9198778728029436,"probe/pre_sft/normal/action_query/current_action/accuracy":0.9455555555555556,"probe/pre_sft/shuffled_frames/dit_global_mean/time_to_next_flap/accuracy":0.8043055555555556,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/first_epoch_loss":1.1951435787621247,"probe/pre_sft/repeated_last_frame/dit_global_mean/time_to_next_flap/macro_f1":0.729707940634347,"probe/post_sft/normal/dit_global_mean/time_since_last_flap/final_epoch_loss":0.09891631855867325,"probe/pre_sft/normal/action_query/time_since_last_flap/first_epoch_loss":1.3115629811785114,"probe/pre_sft/shuffled_frames/action_query/time_since_last_flap/final_epoch_loss":0.6982359878843138,"probe/pre_sft/normal/dit_global_mean/time_to_next_flap/macro_f1":0.7308160252349626,"probe/pre_sft/normal/vae_temporal_delta/time_to_next_flap/macro_f1":0.44813666795670903,"probe/pre_sft/normal/dit_temporal_delta/current_action/accuracy":0.9677777777777777,"report_path":"/workspace/starVLA/results/linear_probe/wan_oft_flappy_latency2_context5_v0_5train_2val/report.json","probe/pre_sft/latency_neutral_prompt/direct_action/accuracy":0.5455555555555556,"probe/post_sft/normal/dit_temporal_delta/time_since_last_flap/macro_f1":0.8899588223668516,"probe/pre_sft/repeated_last_frame/dit_global_mean/current_action/balanced_accuracy":0.953863442242539,"probe/pre_sft/latency_neutral_prompt/dit_global_mean/time_to_next_flap/final_epoch_loss":0.2628381886798316,"probe/post_sft/normal/vae_temporal_delta/time_since_last_flap/balanced_accuracy":0.6123481006820437,"probe/pre_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/final_epoch_loss":0.9897070578360788,"probe/post_sft/shuffled_frames/train_samples":9112,"probe/post_sft/repeated_last_frame/direct_action/balanced_accuracy":0.7773944849574266,"probe/pre_sft/shuffled_frames/action_query/current_action/balanced_accuracy":0.8895758149546982,"probe/pre_sft/latency_neutral_prompt/dit_temporal_delta/current_action/balanced_accuracy":0.9581142678425265,"probe/post_sft/repeated_last_frame/validation_samples":7200,"probe/pre_sft/latency_neutral_prompt/action_query/time_to_next_flap/macro_f1":0.649934085774593,"probe/post_sft/shuffled_frames/vae_temporal_delta/time_to_next_flap/accuracy":0.63625,"probe/post_sft/normal/vae_temporal_delta/current_action/accuracy":0.8505555555555555,"probe/post_sft/shuffled_frames/dit_temporal_delta/current_action/first_epoch_loss":0.8357743695182064,"probe/post_sft/repeated_last_frame/dit_global_mean/current_action/final_epoch_loss":0.10612871049133908,"probe/pre_sft/repeated_last_frame/vae_temporal_delta/time_to_next_flap/macro_f1":0.4411802509773085,"probe/pre_sft/normal/dit_temporal_delta/time_since_last_flap/final_epoch_loss":0.07572646234985816,"probe/post_sft/normal/dit_temporal_delta/current_action/macro_f1":0.9820129421164105,"probe/post_sft/latency_neutral_prompt/vae_temporal_delta/time_since_last_flap/first_epoch_loss":1.5832231149430647,"probe/post_sft/latency_neutral_prompt/dit_global_mean/current_action/accuracy":0.9886111111111111,"probe/post_sft/normal/dit_global_mean/current_action/final_epoch_loss":0.021078848704381717,"probe/post_sft/normal/dit_global_mean/current_action/accuracy":0.9884722222222222} \ No newline at end of file diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb_provenance.json b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb_provenance.json new file mode 100644 index 0000000000000000000000000000000000000000..df86715f91a05e5af5398c07b615c3def6ed1713 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/provenance/wandb_provenance.json @@ -0,0 +1,7411 @@ +[ + { + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "runs": [ + { + "id": "3kohuetr", + "name": "wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-107-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-29T14:50:13.940444Z", + "args": [ + "--config-name", + "train", + "model=wan_oft", + "env=flappy", + "init=wan_oft_libero", + "mode=single", + "++framework.qwenvl.base_vlm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers", + "++framework.qwenvl.attn_implementation=sdpa", + "++framework.qwenvl.enable_gradient_checkpointing=true", + "++framework.action_model.state_dim=7", + "++framework.action_model.loss_type=current_discrete_ce", + "++framework.action_model.action_horizon=8", + "++framework.action_model.future_action_window_size=7", + "++framework.action_model.past_action_window_size=0", + "++framework.action_model.action_model_type=MLP", + "++framework.action_model.action_dim=7", + "++framework.action_model.action_env_dim=2", + "++framework.action_model.action_query_source=mean", + "++framework.action_model.action_query_num_heads=24", + "++framework.name=WanOFT", + "++framework.world_model.base_wm=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers", + "++framework.world_model.extract_layers=[-1]", + "++framework.world_model.num_frames=5", + "++datasets.vla_data.dataset_py=lerobot_datasets", + "++datasets.vla_data.include_state=true", + "++datasets.vla_data.data_root_dir=playground/Datasets/rl_games", + "++datasets.vla_data.data_mix=flappy_train__bridge", + "++datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "++datasets.vla_data.custom_mixtures_path=null", + "++datasets.vla_data.action_type=discrete", + "++datasets.vla_data.sequential_step_sampling=false", + "++datasets.vla_data.eval_sequential_step_sampling=null", + "++datasets.vla_data.num_workers=8", + "++datasets.vla_data.eval_num_workers=8", + "++datasets.vla_data.prefetch_factor=4", + "++datasets.vla_data.persistent_workers=true", + "++datasets.vla_data.pin_memory=true", + "++datasets.vla_data.shuffle=true", + "++datasets.vla_data.action_balance.enabled=false", + "++datasets.vla_data.action_balance.strategy=balanced_epoch", + "++datasets.vla_data.action_balance.action_key=action_id", + "++datasets.vla_data.action_balance.target_flap_fraction=0.3", + "++datasets.vla_data.action_balance.noop_id=0", + "++datasets.vla_data.action_balance.flap_id=1", + "++datasets.vla_data.latency_curriculum.enabled=false", + "++datasets.vla_data.latency_curriculum.strategy=exclusive", + "++datasets.vla_data.latency_curriculum.latencies=null", + "++datasets.vla_data.latency_curriculum.phase_steps=null", + "++datasets.vla_data.latency_curriculum.new_latency_passes=1.0", + "++datasets.vla_data.latency_curriculum.replay_passes=0.25", + "++datasets.vla_data.latency_curriculum.target_total_passes=2.0", + "++datasets.vla_data.latency_curriculum.final_equalization=true", + "++datasets.vla_data.latency_curriculum.step_budget_mode=auto", + "++datasets.vla_data.latency_curriculum.eval_at_phase_end=false", + "++datasets.vla_data.latency_curriculum.save_at_phase_end=false", + "++datasets.vla_data.latency_curriculum.computed_plan=null", + "++datasets.vla_data.latency_curriculum.effective_batch_size=null", + "++datasets.vla_data.per_device_batch_size=4", + "++datasets.vla_data.load_all_data_for_training=true", + "++datasets.vla_data.num_obs_frames=1", + "++datasets.vla_data.image_mode=single", + "++datasets.vla_data.stitch_grid=[2,2]", + "++datasets.vla_data.obs_image_size=[224,224]", + "++datasets.vla_data.video_backend=torchvision_av", + "++datasets.vla_data.observation_indices=[-4,-3,-2,-1,0]", + "++datasets.vla_data.language_indices=[0]", + "++datasets.vla_data.state_indices=[0]", + "++datasets.vla_data.action_indices=[0,1,2,3,4,5,6,7]", + "++datasets.vla_data.pack_image_sequence=true", + "++datasets.vla_data.image_sequence_length=5", + "++datasets.vla_data.context_images_column=observation.context_images", + "++dataset.source_hf=", + "++dataset.config_name=null", + "++dataset.source_subdir=null", + "++dataset.converted_name=flappy_train__bridge", + "++dataset.single_source_hf=", + "++dataset.mixed_source_hf=", + "++dataset.single_converted_name=flappy_train__bridge", + "++dataset.mixed_converted_name=flappy_mixed_latency_train", + "++dataset.single_latency_filter=null", + "++dataset.mixed_latency_filter=null", + "++dataset.force_download=false", + "++dataset.setup_force=false", + "++dataset.skip_verification=false", + "++dataset.verify_rows=200", + "++dataset.max_episodes=null", + "++dataset.episodes_per_latency=null", + "++dataset.latency_filter=null", + "++dataset.debug_subset.enabled=false", + "++dataset.debug_subset.max_episodes=5", + "++dataset.debug_subset.suffix=debug", + "++base_model.repo_id=Wan-AI/Wan2.2-TI2V-5B-Diffusers", + "++initialization.checkpoint_local_dir=playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1", + "++initialization.checkpoint_hf_repo_id=StarVLA/WM4A-Wan2d2-OFT-LIBERO-4in1", + "++initialization.checkpoint_filename=checkpoints/steps_60000_pytorch_model.pt", + "++trainer.max_train_steps=2000", + "++trainer.num_warmup_steps=0", + "++trainer.save_interval=0", + "++trainer.eval_interval=500", + "++trainer.eval_num_batches=50", + "++trainer.per_latency_eval_num_batches=null", + "++trainer.eval_action_classification=false", + "++trainer.eval_action_classification_interval=null", + "++trainer.cc_f1_tolerance=1", + "++trainer.learning_rate.base=2.5e-05", + "++trainer.learning_rate.qwen_vl_interface=1e-05", + "++trainer.learning_rate.action_model=0.0001", + "++trainer.learning_rate.action_query_proj=0.0001", + "++trainer.lr_scheduler_type=cosine_with_min_lr", + "++trainer.scheduler_specific_kwargs.min_lr=1e-06", + "++trainer.freeze_modules=", + "++trainer.freeze_vit=false", + "++trainer.freeze_tied_embedding=false", + "++trainer.freeze_llm_layers=[]", + "++trainer.loss_scale.vla=1.0", + "++trainer.loss_scale.vlm=0.1", + "++trainer.max_grad_norm=1.0", + "++trainer.weight_decay=0.0", + "++trainer.logging_frequency=1", + "++trainer.profile_timing.enabled=false", + "++trainer.profile_timing.log_interval=10", + "++trainer.gradient_clipping=1.0", + "++trainer.gradient_accumulation_steps=32", + "++trainer.distributed_backend=none", + "++trainer.is_resume=false", + "++trainer.pretrained_checkpoint=null", + "++trainer.resume_step=0", + "++trainer.reload_modules=null", + "++trainer.optimizer.name=AdamW", + "++trainer.optimizer.betas=[0.9,0.95]", + "++trainer.optimizer.eps=1e-08", + "++trainer.optimizer.weight_decay=1e-08", + "++trainer.optimizer.fused=true", + "++trainer.save_format=pt", + "++workspace_dir=WORKSPACE_DIR", + "++run_root_dir=results/Checkpoints", + "++seed=42", + "++auth.env_file=null", + "++auth.hf_token_env=HF_TOKEN", + "++auth.wandb_api_key_env=WANDB_API_KEY", + "++paths.run_root_dir=results/Checkpoints", + "++paths.dataset_local_dir=data/flappy_fix_latency_1_200ep_context5", + "++paths.dataset_cache_dir=null", + "++paths.base_model_dir=playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers", + "++paths.accelerate_config=starVLA/config/deepseeds/deepspeed_zero2.yaml", + "++rl_games.model_alias=wan_oft", + "++rl_games.env_eval.image_size=224", + "++rl_games.env_eval.frameskip=1", + "++rl_games.env_eval.seed=42", + "++rl_games.env_eval.fixed_episode_seeds=true", + "++rl_games.env_eval.latency_seed_stride=0", + "++rl_games.env_eval.task_seed_stride=0", + "++rl_games.env_eval.task_description=", + "++rl_games.env_eval.action_chunk_execution.enabled=false", + "++rl_games.env_eval.action_chunk_execution.chunk_size=null", + "++rl_games.env_eval.enabled=true", + "++rl_games.env_eval.distributed_mode=none", + "++rl_games.env_eval.vectorized.enabled=false", + "++rl_games.env_eval.vectorized.batch_size=1", + "++rl_games.env_eval.latency.prompt_map_path=data/flappy_fix_latency_1_200ep_context5/flappy_train__bridge/latency_prompt_map_eval_0_4.json", + "++rl_games.env_eval.latency.mode=single", + "++rl_games.env_eval.latency.values=[0]", + "++rl_games.env_eval.mid_train.enabled=true", + "++rl_games.env_eval.mid_train.interval_steps=500", + "++rl_games.env_eval.mid_train.latencies=[1]", + "++rl_games.env_eval.mid_train.num_episodes=5", + "++rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "++rl_games.env_eval.post_train.enabled=true", + "++rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "++rl_games.env_eval.post_train.num_episodes=20", + "++rl_games.env_eval.post_train.max_steps_per_episode=3600", + "++rl_games.task=flappy", + "++rl_games.initialization_mode=bridge", + "++rl_games.action_carrier=bridge", + "++checkpoint.load=none", + "++checkpoint.hf_repo_id=null", + "++checkpoint.save_best_model=false", + "++checkpoint.save_final_model=true", + "++checkpoint.save_pt_file=true", + "++checkpoint.save_training_state=false", + "++checkpoint.local.keep_last_n=1", + "++checkpoint.sync.enabled=false", + "++checkpoint.sync.repo_id=null", + "++checkpoint.sync.keep_last_n=0", + "++checkpoint.sync.sync_every_n_checkpoints=1", + "++checkpoint.sync.resume_policy=local_latest", + "++run_id=wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce", + "++output_dir=null", + "++config_yaml=null", + "++is_debug=false", + "++version_id=0.21", + "++run_root_dir=/workspace/starVLA/results/Checkpoints", + "++trainer.is_resume=false", + "++trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/WM4A-Wan2d2-OFT-LIBERO-4in1/checkpoints/steps_60000_pytorch_model.pt", + "++trainer.resume_step=0", + "++datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1_200ep_context5", + "++framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers", + "++framework.world_model.base_wm=/workspace/starVLA/playground/Pretrained_models/Wan-AI/Wan2.2-TI2V-5B-Diffusers" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce/wandb", + "host": "e2951ec2d005", + "executable": "/venv/starvla_wanoft/bin/python", + "cpu_count": 16, + "cpu_count_logical": 32, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "214748364800", + "used": "124962414592" + } + }, + "memory": { + "total": "200128028672" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-94ca9dbe-3c07-4bb5-918b-9c38a98d905b" + } + ], + "cudaVersion": "13.2", + "writerId": "rgp0wmuz67h6bfhtzlk4y156pre6t34n" + } + }, + { + "id": "del3lyp2", + "name": "flappy_curriculum_exclusive_40ep_per_latency", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-181-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-20T18:47:13.132069Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=curriculum_exclusive", + "run_id=flappy_curriculum_exclusive_40ep_per_latency", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/flappy_curriculum_exclusive_40ep_per_latency/checkpoints/steps_4100_state", + "trainer.resume_step=4100", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=true", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=true", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.latency_curriculum.latencies=[0,1,2,3,4]", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat", + "datasets.vla_data.data_mix=flappy_mixed_latency_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_mixed_latency_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=mixed", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0,1,2,3,4]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat/flappy_mixed_latency_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0,1,2,3,4,5]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "ba36a6b93440808752b66ba5d25281b4c0ce5a8b" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_curriculum_exclusive_40ep_per_latency/wandb", + "host": "bd9fab9545e8", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 64, + "cpu_count_logical": 128, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "109846835200" + } + }, + "memory": { + "total": "405145010176" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-6bd86989-cb7c-f5ab-5feb-b2b65ec080f1" + } + ], + "cudaVersion": "13.2", + "writerId": "islw1fggur9gfwsp4x4nylgrj7f6korc" + } + }, + { + "id": "n1fo4bsj", + "name": "flappy_curriculum_exclusive_40ep_per_latency", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-181-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-20T11:55:05.120175Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=curriculum_exclusive", + "run_id=flappy_curriculum_exclusive_40ep_per_latency", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/flappy_curriculum_exclusive_40ep_per_latency/checkpoints/steps_2600_state", + "trainer.resume_step=2600", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=true", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=true", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.latency_curriculum.latencies=[0,1,2,3,4,5]", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat", + "datasets.vla_data.data_mix=flappy_mixed_latency_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_mixed_latency_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=mixed", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0,1,2,3,4,5]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat/flappy_mixed_latency_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0,1,2,3,4,5]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "ba36a6b93440808752b66ba5d25281b4c0ce5a8b" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_curriculum_exclusive_40ep_per_latency/wandb", + "host": "bd9fab9545e8", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 64, + "cpu_count_logical": 128, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "109797208064" + } + }, + "memory": { + "total": "405145010176" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-6bd86989-cb7c-f5ab-5feb-b2b65ec080f1" + } + ], + "cudaVersion": "13.2", + "writerId": "qlvsp00evk6mph90zkwlhqwxo523sj5y" + } + }, + { + "id": "0yhpy1gn", + "name": "flappy_curriculum_exclusive_40ep_per_latency", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-19T15:32:39.996135Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=curriculum_exclusive", + "run_id=flappy_curriculum_exclusive_40ep_per_latency", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=true", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=true", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.latency_curriculum.latencies=[0,1,2,3,4,5]", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat", + "datasets.vla_data.data_mix=flappy_mixed_latency_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_mixed_latency_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=mixed", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0,1,2,3,4,5]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat/flappy_mixed_latency_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0,1,2,3,4,5]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "d1a786ff88f4a5aff08baa76b8c82ab4964a2804" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_curriculum_exclusive_40ep_per_latency/wandb", + "host": "9ecb63543e57", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "90273775616" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b39e8721-119a-1ff6-40d5-a3b2d9a201ed" + } + ], + "cudaVersion": "13.1", + "writerId": "s9ke158w4la7j6u4qq7og53uzh69a0hz" + } + }, + { + "id": "il8v7kn6", + "name": "flappy_curriculum_cumulative_40ep_per_latency", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-124-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-19T09:49:02.828635Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=curriculum_cumulative", + "run_id=flappy_curriculum_cumulative_40ep_per_latency", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=true", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=true", + "datasets.vla_data.latency_curriculum.strategy=cumulative", + "datasets.vla_data.latency_curriculum.latencies=[0,1,2,3,4,5]", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat", + "datasets.vla_data.data_mix=flappy_mixed_latency_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_mixed_latency_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=mixed", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0,1,2,3,4,5]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_mixed_latency_40ep_per_lat/flappy_mixed_latency_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0,1,2,3,4,5]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "d1a786ff88f4a5aff08baa76b8c82ab4964a2804" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_curriculum_cumulative_40ep_per_latency/wandb", + "host": "df573cb221ea", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 16, + "cpu_count_logical": 24, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "121535139840" + } + }, + "memory": { + "total": "134886748160" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-512aa42b-1025-ef3b-2393-04d299de42e5" + } + ], + "cudaVersion": "13.2", + "writerId": "f2ydq1teg02leeppdg1k5qhbndaaqmbz" + } + }, + { + "id": "j8qt7y5y", + "name": "demon_attack_fix_latency_4_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-119-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "started_at": "2026-06-17T16:19:24.157074Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=demon_attack_fix_latency_4_200ep", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=7000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=false", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_4_200ep", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_4_200ep/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/demon_attack_fix_latency_4_200ep/wandb", + "host": "stair--833ef67bd2ee-bozqx7lc4x", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "writer_id": "fjztld68uibv4sgxlagek6jgor2qr29i" + } + }, + { + "id": "fhar1jmq", + "name": "demon_attack_fix_latency_2_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-119-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "started_at": "2026-06-17T16:16:35.756274Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=demon_attack_fix_latency_2_200ep", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=7000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=false", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_2_200ep", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_2_200ep/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/demon_attack_fix_latency_2_200ep/wandb", + "host": "stair--833ef67bd2ee-bozqx7lc4x", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "writer_id": "omtq98gg00y2mp1zz86g8n38nut163bu" + } + }, + { + "id": "auq6wt48", + "name": "demon_attack_fix_latency_8_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-89-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-17T16:49:19.701833Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=demon_attack_fix_latency_8_200ep", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=7000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=false", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_8_200ep", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_8_200ep/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/demon_attack_fix_latency_8_200ep/wandb", + "host": "stair-exp--c9b59aa83605-uyar5wadi3", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "writer_id": "825yeddydai1yuxt0k78iuhu3lcv00c9" + } + }, + { + "id": "k4pvtjp3", + "name": "demon_attack_fix_latency_6_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-89-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-17T16:49:19.701671Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=demon_attack_fix_latency_6_200ep", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=7000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=false", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_6_200ep", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/demon_attack_fix_latency_6_200ep/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "dfb71eef10b42a14f04f09b297665fcbb6cf77bd" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/demon_attack_fix_latency_6_200ep/wandb", + "host": "stair-exp--c9b59aa83605-uyar5wadi3", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "writer_id": "srnatk6zncucd87ixidttbvsn8j3c2jq" + } + }, + { + "id": "x1fvp0qj", + "name": "demon_attack_fix_latency_0_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-124-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-17T06:12:42.075475Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=demon_attack_fix_latency_0_200ep", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=7000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9,0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.latency_curriculum.enabled=false", + "datasets.vla_data.latency_curriculum.strategy=exclusive", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/demon_attack_fix_latency_0_200ep", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/demon_attack_fix_latency_0_200ep/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=500", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "1a93a1ae6772e74087a836b7b2246d340b948bd2" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/demon_attack_fix_latency_0_200ep/wandb", + "host": "df573cb221ea", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 16, + "cpu_count_logical": 24, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "66937421824" + } + }, + "memory": { + "total": "134886748160" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-512aa42b-1025-ef3b-2393-04d299de42e5" + } + ], + "cudaVersion": "13.2", + "writerId": "nj74gwpul6uqopnch1b9lrnmq4sgabnq" + } + }, + { + "id": "i2k3ascz", + "name": "flappy_fix_latency_3_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-49-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-15T19:05:33.451159Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=flappy_fix_latency_3_200ep", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_3_200ep", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "ffe48f30f636bd6984f9d8c37e9b54c9f474c172" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_fix_latency_3_200ep/wandb", + "host": "6ad90cfaa540", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "44202926080" + } + }, + "memory": { + "total": "1081993768960" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-559b7ac2-9d73-1ff1-1c3d-dd7aa0f27221" + } + ], + "cudaVersion": "13.1", + "writerId": "7tgyoxxo1wn6qdpsedu2wke1as3wvivr" + } + }, + { + "id": "px17hbgd", + "name": "flappy_fix_latency_4_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-124-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-15T19:00:56.812629Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=flappy_fix_latency_4_200ep", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_4_200ep", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "ffe48f30f636bd6984f9d8c37e9b54c9f474c172" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_fix_latency_4_200ep/wandb", + "host": "df573cb221ea", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 16, + "cpu_count_logical": 24, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "44019662848" + } + }, + "memory": { + "total": "134886748160" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-512aa42b-1025-ef3b-2393-04d299de42e5" + } + ], + "cudaVersion": "13.2", + "writerId": "t3vpeff79umhns1sl2ens65qul9eivxl" + } + }, + { + "id": "zdoydsz5", + "name": "flappy_fix_latency_2_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-15T18:46:37.908727Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=flappy_fix_latency_2_200ep", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2_200ep", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "ffe48f30f636bd6984f9d8c37e9b54c9f474c172" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_fix_latency_2_200ep/wandb", + "host": "1d9209ed901f", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "44738023424" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-0ba3d5a5-b286-0c1e-15a8-acb0a171b2c7" + } + ], + "cudaVersion": "13.1", + "writerId": "9nnaxjwjgxc3yuw9bbnkdt94hzdf7oiy" + } + }, + { + "id": "hexy6iks", + "name": "flappy_fix_latency_0_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-58-lowlatency-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-15T17:44:47.048650Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=flappy_fix_latency_0_200ep", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/flappy_fix_latency_0_200ep/checkpoints/steps_1300_state", + "trainer.resume_step=1300", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_0_200ep", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fa2cf48737a4e7c4942bb9ac8d6d2e52880fb456" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_fix_latency_0_200ep/wandb", + "host": "89df82bc5f93", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "70385577984" + } + }, + "memory": { + "total": "1081996767232" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-0de8ee2f-463a-3b51-2174-ef9f8b3c06b0" + } + ], + "cudaVersion": "13.1", + "writerId": "lqik1edpzvbdwg8as5potv01u8lsw58s" + } + }, + { + "id": "8d9umgvm", + "name": "flappy_fix_latency_1_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.11.0-1027-oem-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-15T17:11:58.056740Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=flappy_fix_latency_1_200ep", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1_200ep", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "d05200d75af235edfb5db5bccd1f870625e288c1" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_fix_latency_1_200ep/wandb", + "host": "cc9891052b95", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "43569098752" + } + }, + "memory": { + "total": "1352265048064" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-a32efffc-8eb9-fb3e-b951-6f8cea2edd91" + } + ], + "cudaVersion": "13.1", + "writerId": "wabhxaptcmrdw2pw0hnm0pj2czj1t1wv" + } + }, + { + "id": "qqbbhk6o", + "name": "flappy_fix_latency_0_200ep", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-58-lowlatency-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-15T13:10:08.888803Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=flappy_fix_latency_0_200ep", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=false", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=100", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=4", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.shuffle=true", + "datasets.vla_data.action_balance.enabled=false", + "datasets.vla_data.action_balance.strategy=balanced_epoch", + "datasets.vla_data.action_balance.action_key=action_id", + "datasets.vla_data.action_balance.target_flap_fraction=0.3", + "datasets.vla_data.action_balance.noop_id=0", + "datasets.vla_data.action_balance.flap_id=1", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_0_200ep", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fa2cf48737a4e7c4942bb9ac8d6d2e52880fb456" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/flappy_fix_latency_0_200ep/wandb", + "host": "89df82bc5f93", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "42963836928" + } + }, + "memory": { + "total": "1081996767232" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-0de8ee2f-463a-3b51-2174-ef9f8b3c06b0" + } + ], + "cudaVersion": "13.1", + "writerId": "jh37ewptg9frb5byb06svf9c3dohclc3" + } + }, + { + "id": "dd7te8ee", + "name": "openvla_deadly_corridor_fix_latency_2", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-89-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-10T15:22:35.301231Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_2", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=50", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/deadly_corridor_fix_latency_2", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/deadly_corridor_fix_latency_2/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=50", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,2,4,6,8]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/openvla_deadly_corridor_fix_latency_2/wandb", + "host": "stair-exp--c9b59aa83605-gw2mg67ela", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu_type": "NVIDIA H200", + "gpu_count": 2, + "disk": { + "/": { + "total": "3838880616448", + "used": "2935390556160" + } + }, + "memory": { + "total": "2163843989504" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-796ff422-81d6-0212-5213-a4f446e849fe" + }, + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-b86d9a9c-71ca-67e2-08fa-e881b7d278cf" + } + ], + "cuda_version": "12.8", + "writer_id": "x619mwwkz58s8nc3l6vdoga95asyropp" + } + }, + { + "id": "bqlkwsrg", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-89-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-10T15:23:14.264574Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=50", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=1", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=50", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,2,4,6,8]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "stair-exp--c9b59aa83605-gw2mg67ela", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu_type": "NVIDIA H200", + "gpu_count": 2, + "disk": { + "/": { + "total": "3838880616448", + "used": "2935390474240" + } + }, + "memory": { + "total": "2163843989504" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-796ff422-81d6-0212-5213-a4f446e849fe" + }, + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-b86d9a9c-71ca-67e2-08fa-e881b7d278cf" + } + ], + "cuda_version": "12.8", + "writer_id": "8zgyahmjwzylbiupgwae2wig8f1asymh" + } + }, + { + "id": "i94788vg", + "name": "openvla_flappy_fix_latency_2_small", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "started_at": "2026-06-11T08:03:37.824268Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2_small", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2_small/checkpoints/steps_1500_state", + "trainer.resume_step=1500", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2_small", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_2_small/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "d182480d774978c8e37efb1d3ea983d812acd4f6" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2_small/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu_type": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "128877953024" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memory_total": "102641958912", + "cuda_cores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cuda_version": "13.1", + "writer_id": "plfydg31coo2ac4yenpxk14lm7zy6lez" + } + }, + { + "id": "p6sf7031", + "name": "openvla_flappy_fix_latency_4_small", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-119-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-09T01:47:54.705010Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_4_small", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_4_small", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_4_small/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[4]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/openvla_flappy_fix_latency_4_small/wandb", + "host": "stair-exp-0--bf9a78781a26-msepuytk2l", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu_type": "NVIDIA H200", + "gpu_count": 1, + "disk": { + "/": { + "total": "3838880616448", + "used": "2165519695872" + } + }, + "memory": { + "total": "2164119814144" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-ae5cbab9-107e-1030-b8ed-86ab3ed49518" + } + ], + "cuda_version": "12.8", + "writer_id": "t1oowdbbvk57osz7pgo61q38aiwd7xu9" + } + }, + { + "id": "4oyyy1lw", + "name": "openvla_flappy_fix_latency_3_small", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-89-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-08T17:39:12.924292Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_3_small", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_3_small", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_3_small/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[3]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/openvla_flappy_fix_latency_3_small/wandb", + "host": "stair-exp--c9b59aa83605-bqvx6cydwg", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu_type": "NVIDIA H200", + "gpu_count": 2, + "disk": { + "/": { + "total": "3838880616448", + "used": "2376642682880" + } + }, + "memory": { + "total": "2163841486848" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-fbbfd2f5-0cb0-cbc4-d109-eec4a153ca50" + }, + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-f34038f4-74bb-775d-6d36-52ce678310a6" + } + ], + "cuda_version": "12.8", + "writer_id": "dy8dnf223zi1p1v3sqvg5uqtssvmkq4j" + } + }, + { + "id": "kttbna75", + "name": "openvla_flappy_fix_latency_2_small", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-89-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-08T17:28:42.534945Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2_small", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_2_small", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_2_small/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/openvla_flappy_fix_latency_2_small/wandb", + "host": "stair-exp--c9b59aa83605-bqvx6cydwg", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu_type": "NVIDIA H200", + "gpu_count": 2, + "disk": { + "/": { + "total": "3838880616448", + "used": "2376641536000" + } + }, + "memory": { + "total": "2163841486848" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-fbbfd2f5-0cb0-cbc4-d109-eec4a153ca50" + }, + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-f34038f4-74bb-775d-6d36-52ce678310a6" + } + ], + "cuda_version": "12.8", + "writer_id": "det3s55lrfyxmj6058o2zmzqvzg0jxgt" + } + }, + { + "id": "kn1uukiy", + "name": "openvla_flappy_fix_latency_0_small", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-119-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "started_at": "2026-06-08T15:40:41.427091Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_0_small", + "run_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_0_small", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/data/flappy_fix_latency_0_small/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/inspire/hdd/project/spatialintelligence/public/lzj/starVLA/starVLA/training/train_starvla_hydra.py", + "code_path": "starVLA/training/train_starvla_hydra.py", + "code_path_local": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote_url": "https://github.com/talha1503/starVLA.git", + "commit": "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + }, + "root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/starvla/results/Checkpoints/openvla_flappy_fix_latency_0_small/wandb", + "host": "stair-exp-0--bf9a78781a26-msepuytk2l", + "executable": "/inspire/hdd/project/spatialintelligence/public/lzj/miniconda3/envs/starvla_openvla/bin/python", + "cpu_count": 96, + "cpu_count_logical": 192, + "gpu_type": "NVIDIA H200", + "gpu_count": 1, + "disk": { + "/": { + "total": "3838880616448", + "used": "2165508112384" + } + }, + "memory": { + "total": "2164119814144" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA H200", + "memory_total": "150754820096", + "cuda_cores": 16896, + "architecture": "Hopper", + "uuid": "GPU-ae5cbab9-107e-1030-b8ed-86ab3ed49518" + } + ], + "cuda_version": "12.8", + "writer_id": "83qo07a0e4z4nbpfvfu3irgkn15e0v9i" + } + }, + { + "id": "l1gtf7dc", + "name": "openvla_flappy_fix_latency_1_small", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-08T09:52:48.491301Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1_tmp", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1_tmp", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1_tmp/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[1]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "1b57b562a6c2ffb2fd549282e0baaa1f86661123" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1_tmp/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "72050561024" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "gleq31s520fss6antjdn8z593kw3mxd8" + } + }, + { + "id": "lftcj17d", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-08T02:28:40.752684Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/checkpoints/steps_500_state", + "trainer.resume_step=500", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[1]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "0bc71a07a07274d35bc9aa60f8d8d14306d45c24" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "123950485504" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "q3pdk48pkts0gntyrorz6d7ezf4isll0" + } + }, + { + "id": "zhon9num", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-07T18:57:00.062558Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/checkpoints/steps_400_state", + "trainer.resume_step=400", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[1]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "0bc71a07a07274d35bc9aa60f8d8d14306d45c24" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "123949047808" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "44q16xdxvavzayksibbej91m3csw88kd" + } + }, + { + "id": "6s9ihina", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-07T16:32:20.634463Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[1]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "0bc71a07a07274d35bc9aa60f8d8d14306d45c24" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "69115834368" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "ao6joc9pia8et81ejfflshlsjodw1w6c" + } + }, + { + "id": "ef6e3dz3", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-07T16:30:58.648116Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.distributed_mode=none", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[1]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "0bc71a07a07274d35bc9aa60f8d8d14306d45c24" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "69115858944" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "rhxgvu8eyjiqv9yk50aggi440ai2lafp" + } + }, + { + "id": "lalgz7p2", + "name": "openvla_flappy_fix_latency_2", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-07T09:38:13.417721Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/checkpoints/best_state", + "trainer.resume_step=200", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_2/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "95003144192" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "h801cta0cpudm4ekih0kiv0miu37e7b1" + } + }, + { + "id": "6eauedlc", + "name": "openvla_flappy_fix_latency_2", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-07T08:44:20.799750Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/checkpoints/best_state", + "trainer.resume_step=200", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_2/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "95002910720" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "goag0qjd4tz8nrb26s6q6vyby6t9l6ie" + } + }, + { + "id": "5hkk2h4g", + "name": "openvla_flappy_fix_latency_2", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-07T07:01:54.955810Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=0", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_2/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "67585515520" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "g9xhnuahkpc97vdki3om217y7nan9c5g" + } + }, + { + "id": "bry8it5y", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-06T17:46:17.660442Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[1]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "57541623808" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "4lzp764q6h0wdb1bm8icexu84uzut1nd" + } + }, + { + "id": "wwj8ckn4", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-06T08:48:11.268338Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/checkpoints/best_state", + "trainer.resume_step=100", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "73989873664" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "6by1q1n71o0cxdgl12rmsqufkholul3d" + } + }, + { + "id": "wyixw7oe", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-06T08:47:04.015045Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/checkpoints/best_state", + "trainer.resume_step=100", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "73989795840" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "6qyi3p6j3dwxvxxrx3qaxuikewh1cts3" + } + }, + { + "id": "bfsickej", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-06T08:45:56.734756Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "73989718016" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "azmbvu8ruqvrxjsv9u37ok8e1n3n4uj0" + } + }, + { + "id": "r15xbzb8", + "name": "openvla_flappy_fix_latency_1", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-06T06:56:47.187409Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_1", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=1500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_1", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_1/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "fc2ae1e2d04ca9ad021b9b46dfd0a2ee3144bc9e" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_1/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "80574586880" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "hsyamhwrfwpe8xqdjef0pvddacknu882" + } + }, + { + "id": "5jkdo5rb", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-05T04:11:05.908069Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/checkpoints/best_state", + "trainer.resume_step=800", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "9e1cf489a76ed2c07abc29399d637d5d73417ccb" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "90910617600" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "pecxv0il5ocpucg92geyg32v9q3kf4bn" + } + }, + { + "id": "m9wzp52s", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-05T04:10:01.335390Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/checkpoints/best_state", + "trainer.resume_step=800", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "9e1cf489a76ed2c07abc29399d637d5d73417ccb" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "90910478336" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "cx1a2xhrg4mnzw145t7zfsmxzpe2vp3z" + } + }, + { + "id": "u6872md1", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-05T03:25:54.690898Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "checkpoint.save_pt_file=false", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "914436f584ae969603a7e9b617a400f266e7fb9f" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "90910281728" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "42aqotcobm8n8v3i4vtrig4gykctu6rs" + } + }, + { + "id": "255dpkdc", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-05T00:05:50.459905Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/checkpoints/best_state", + "trainer.resume_step=800", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "917d564a90c8209144f9b50204c47192f27e1ac1" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "90907291648" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "9bowkst1bklk76w1fb9sswcvk8d6jpkf" + } + }, + { + "id": "a7tq8mj3", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-04T04:26:04.624714Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "917d564a90c8209144f9b50204c47192f27e1ac1" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "63446970368" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "72bxyae4es71o0kvom7pgq0c4hnpzvbq" + } + }, + { + "id": "4dwv1okj", + "name": "openvla_demon_attack_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-04T04:17:31.681712Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=openvla_demon_attack_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_demon_attack_fix_latency_0/checkpoints/best_state", + "trainer.resume_step=800", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/demon_attack_fix_latency_0", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/demon_attack_fix_latency_0/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "917d564a90c8209144f9b50204c47192f27e1ac1" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_demon_attack_fix_latency_0/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "70838284288" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "2filbo3igg8l1mtz9728bnoiemm8je8b" + } + }, + { + "id": "fnvodu2y", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.17.0-23-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-03T13:00:22.604920Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "trainer.is_resume=true", + "trainer.pretrained_checkpoint=/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/checkpoints/best_state", + "trainer.resume_step=400", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "917d564a90c8209144f9b50204c47192f27e1ac1" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "79a2aa03e7d1", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 64, + "cpu_count_logical": 128, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "64760532992" + } + }, + "memory": { + "total": "1622997454848" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-28dc48f6-1588-b829-4cd0-1471aa861092" + } + ], + "cudaVersion": "13.2", + "writerId": "nt1eke8yqluqp4bnt7zqqhgbn4ussocb" + } + }, + { + "id": "9cmg0kv4", + "name": "openvla_demon_attack_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-03T12:58:38.676533Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=openvla_demon_attack_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/demon_attack_fix_latency_0", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=0", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/demon_attack_fix_latency_0/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "917d564a90c8209144f9b50204c47192f27e1ac1" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_demon_attack_fix_latency_0/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "42835193856" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "x197e5ik1712dfr5qd3jrw95exmu5d26" + } + }, + { + "id": "1gbmlxx5", + "name": "openvla_demon_attack_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-90-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-03T12:55:39.898970Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=demon_attack", + "init=bridge", + "mode=single", + "run_id=openvla_demon_attack_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=6", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/demon_attack_fix_latency_0", + "datasets.vla_data.data_mix=demon_attack_train__bridge", + "datasets.vla_data.eval_data_mix=demon_attack_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=demon_attack", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=100000", + "rl_games.env_eval.task_description='You are playing Demon Attack from a single game image. Choose exactly one action from: NOOP, FIRE, RIGHT, LEFT, RIGHTFIRE, LEFTFIRE.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/demon_attack_fix_latency_0/demon_attack_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "d9dfe4578bc891fe65984d0daaad1e9783ac51de" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_demon_attack_fix_latency_0/wandb", + "host": "2de6a6c642d9", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 104, + "cpu_count_logical": 208, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "42835079168" + } + }, + "memory": { + "total": "1081782173696" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-5d139805-8a2f-eab7-b26a-9323f90aa0ff" + } + ], + "cudaVersion": "13.1", + "writerId": "blxvljaj18ljtnzd6zj2sk5ti3w79luz" + } + }, + { + "id": "f6q4xd8a", + "name": "openvla_deadly_corridor_fix_latency_0", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.17.0-23-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-03T10:31:49.031040Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=deadly_corridor", + "init=bridge", + "mode=single", + "run_id=openvla_deadly_corridor_fix_latency_0", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "checkpoint.save_best_model=true", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=7", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/deadly_corridor_fix_latency_0", + "datasets.vla_data.data_mix=deadly_corridor_train__bridge", + "datasets.vla_data.eval_data_mix=deadly_corridor_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=deadly_corridor", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=4", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=100000", + "rl_games.env_eval.task_description='You are playing Deadly Corridor in VizDoom. Choose actions from MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, ATTACK.'", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/deadly_corridor_fix_latency_0/deadly_corridor_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[0]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]", + "rl_games.env_eval.deadly.action_layout=multibinary_7" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "d9dfe4578bc891fe65984d0daaad1e9783ac51de" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_deadly_corridor_fix_latency_0/wandb", + "host": "79a2aa03e7d1", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 64, + "cpu_count_logical": 128, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "37342433280" + } + }, + "memory": { + "total": "1622997454848" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-28dc48f6-1588-b829-4cd0-1471aa861092" + } + ], + "cudaVersion": "13.2", + "writerId": "kdts7fwxllfqr1iewwqrax04p96j1rdu" + } + }, + { + "id": "h0s1jx4o", + "name": "openvla_flappy_fix_latency_4_exp", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.17.0-23-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-02T15:21:39.145580Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_4_exp", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=2", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=2500", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_4", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.seed=42", + "rl_games.env_eval.fixed_episode_seeds=true", + "rl_games.env_eval.latency_seed_stride=0", + "rl_games.env_eval.task_seed_stride=100000", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_4/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[4]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "46e36dafd933fd767313c1c84f7ace66c2683be2" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_4_exp/wandb", + "host": "79a2aa03e7d1", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 64, + "cpu_count_logical": 128, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "44940775424" + } + }, + "memory": { + "total": "1622997454848" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-28dc48f6-1588-b829-4cd0-1471aa861092" + } + ], + "cudaVersion": "13.2", + "writerId": "c7d4b4k8fsrtag930evrxqtlstzedlc3" + } + }, + { + "id": "ujfddciy", + "name": "openvla_flappy_fix_latency_6", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.17.0-29-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-05-30T14:20:01.153814Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_6", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=16", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_6", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_6/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=1", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[6]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4,5,6,7]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "1e638dd5cb934c2e322f2f3879174c7d626e4410" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_6/wandb", + "host": "e2aa5dc9d0e7", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Server Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "45404827648" + } + }, + "memory": { + "total": "1216822988800" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Server Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-1a380467-3b9a-a204-dedc-7be14df4c44f" + } + ], + "cudaVersion": "13.2", + "writerId": "f9e2sp6oj4ml3rmcmjpxqn0ahc9rh94r" + } + }, + { + "id": "x7fa6nuy", + "name": "openvla_flappy_fix_latency_2", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-05-30T07:43:47.737646Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=16", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_2/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=1", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4,5,6,7]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "1e638dd5cb934c2e322f2f3879174c7d626e4410" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "60565184512" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "8efyjd78toz3e7rbztx428y6mxu9noq7" + } + }, + { + "id": "dgphne2s", + "name": "openvla_flappy_fix_latency_2", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-05-30T07:40:02.411152Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=16", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_2/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=1", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4,5,6,7]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "1e638dd5cb934c2e322f2f3879174c7d626e4410" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "60565143552" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "jum3piik2cac1m3my66hj4r1ojfl21ze" + } + }, + { + "id": "xiz6nc3h", + "name": "openvla_flappy_fix_latency_2", + "entity": "dongqianyu99-zhejiang-university", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-71-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-05-30T07:32:36.111562Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "init=bridge", + "mode=single", + "run_id=openvla_flappy_fix_latency_2", + "run_root_dir=/workspace/starVLA/results/Checkpoints", + "seed=42", + "wandb_entity=dongqianyu99-zhejiang-university", + "wandb_project=starVLA_rl_games", + "rl_games.env_eval.enabled=true", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "checkpoint.local.keep_last_n=1", + "trainer.is_resume=false", + "trainer.pretrained_checkpoint=/workspace/starVLA/playground/Pretrained_models/Qwen3VL-OFT-Bridge-RT-1/checkpoints/steps_5000_pytorch_model.pt", + "trainer.resume_step=0", + "trainer.max_train_steps=5000", + "trainer.num_warmup_steps=100", + "trainer.save_interval=100", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.logging_frequency=10", + "trainer.gradient_accumulation_steps=16", + "trainer.distributed_backend=none", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=1e-05", + "trainer.learning_rate.action_model=0.0001", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.max_grad_norm=1.0", + "trainer.weight_decay=0.0", + "trainer.gradient_clipping=1.0", + "trainer.optimizer.name=AdamW", + "trainer.optimizer.betas=[0.9, 0.95]", + "trainer.optimizer.eps=1e-08", + "trainer.optimizer.weight_decay=1e-08", + "trainer.optimizer.fused=false", + "trainer.save_format=pt", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.action_model.action_dim=7", + "framework.action_model.action_env_dim=2", + "framework.action_model.state_dim=7", + "framework.action_model.loss_type=discrete_ce", + "framework.action_model.action_horizon=1", + "framework.action_model.future_action_window_size=0", + "framework.action_model.past_action_window_size=0", + "datasets.vla_data.include_state=true", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.per_device_batch_size=16", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.video_backend=torchvision_av", + "datasets.vla_data.data_root_dir=/workspace/starVLA/data/flappy_fix_latency_2", + "datasets.vla_data.data_mix=flappy_train__bridge", + "datasets.vla_data.eval_data_mix=flappy_train__bridge__val", + "framework.qwenvl.base_vlm=/workspace/starVLA/playground/Pretrained_models/Qwen3-VL-4B-Instruct", + "rl_games.task=flappy", + "rl_games.model_alias=openvla", + "rl_games.initialization_mode=bridge", + "rl_games.action_carrier=bridge", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=224", + "rl_games.env_eval.latency.values=[0]", + "rl_games.env_eval.latency.prompt_map_path=/workspace/starVLA/data/flappy_fix_latency_2/flappy_train__bridge/latency_prompt_map.json", + "rl_games.env_eval.mid_train.enabled=true", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.num_episodes=1", + "rl_games.env_eval.mid_train.max_steps_per_episode=3600", + "rl_games.env_eval.mid_train.latencies=[6]", + "rl_games.env_eval.post_train.enabled=true", + "rl_games.env_eval.post_train.num_episodes=5", + "rl_games.env_eval.post_train.max_steps_per_episode=3600", + "rl_games.env_eval.post_train.latencies=[0,1,2,3,4,5,6,7]" + ], + "program": "/workspace/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "https://github.com/talha1503/starVLA.git", + "commit": "1e638dd5cb934c2e322f2f3879174c7d626e4410" + }, + "email": "dongqianyu99@gmail.com", + "root": "/workspace/starVLA/results/Checkpoints/openvla_flappy_fix_latency_2/wandb", + "host": "1430cba38133", + "executable": "/venv/starvla_openvla/bin/python", + "cpu_count": 192, + "cpu_count_logical": 384, + "gpu": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "gpu_count": 1, + "disk": { + "/": { + "total": "161061273600", + "used": "106185515008" + } + }, + "memory": { + "total": "1622828187648" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA RTX PRO 6000 Blackwell Workstation Edition", + "memoryTotal": "102641958912", + "cudaCores": 24064, + "architecture": "Blackwell", + "uuid": "GPU-b8c17823-a25f-beb5-5732-bb224dbaa9c4" + } + ], + "cudaVersion": "13.1", + "writerId": "knhz5b85fjkblsx17ja63w3wn1fyzaes" + } + } + ] + }, + { + "entity": "your_wandb_entity", + "project": "starVLA_rl_games", + "errors": { + "data": { + "project": null + } + } + }, + { + "entity": "zihanwang-ai-northwestern-university", + "project": "starVLA_rl_games", + "runs": [] + }, + { + "entity": "saberrr-zju", + "project": "starVLA_rl_games", + "runs": [ + { + "id": "l6bh6k6d", + "name": "openvla_flappy_l2_full", + "entity": "saberrr-zju", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-100-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-02T19:14:27.680532Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "mode=single", + "run_root_dir=/root/latency-sensitive-bench/results/Checkpoints", + "checkpoint.local.keep_last_n=3", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.obs_image_size=[224,224]", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.video_backend=torchvision_av", + "framework.action_model.action_layout=flappy_categorical_2", + "framework.action_model.loss_type=ce", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.qwenvl.lora.alpha=64", + "framework.qwenvl.lora.bias=none", + "framework.qwenvl.lora.dropout=0.05", + "framework.qwenvl.lora.enabled=false", + "framework.qwenvl.lora.r=32", + "framework.qwenvl.lora.target_modules=all-linear", + "rl_games.env_eval.enabled=false", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=84", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.latency.values=[2]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.mid_train.max_steps_per_episode=2000", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.latencies=[2]", + "rl_games.env_eval.post_train.max_steps_per_episode=2000", + "rl_games.env_eval.post_train.num_episodes=50", + "rl_games.model_alias=openvla", + "rl_games.task=flappy", + "run_id=openvla_flappy_l2_full", + "seed=42", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.gradient_accumulation_steps=2", + "trainer.gradient_clipping=1.0", + "trainer.learning_rate.action_model=0.0001", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=2e-05", + "trainer.logging_frequency=10", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.max_grad_norm=1.0", + "trainer.max_train_steps=2000", + "trainer.num_warmup_steps=100", + "trainer.save_format=pt", + "trainer.save_interval=2001", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.weight_decay=0.0", + "wandb_entity=saberrr-zju", + "wandb_project=starVLA_rl_games", + "datasets.vla_data.data_root_dir=/root/latency-sensitive-bench/playground/Datasets/rl_games", + "datasets.vla_data.data_mix=flappy_fixed_l2_fs4", + "datasets.vla_data.eval_data_mix=flappy_fixed_l2_fs4__val", + "framework.qwenvl.base_vlm=/root/latency-sensitive-bench/playground/Pretrained_models/Qwen3-VL-4B-Instruct-Action", + "rl_games.env_eval.latency.prompt_map_path=/root/latency-sensitive-bench/playground/Datasets/rl_games/flappy_fixed_l2_fs4/latency_prompt_map.json", + "trainer.is_resume=false" + ], + "program": "/root/latency-sensitive-bench/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "git@github.com:talha1503/starVLA.git", + "commit": "538fd523fe4fa4e7abce3201611919314bc6b664" + }, + "email": "lxy1501454486@gmail.com", + "root": "/root/latency-sensitive-bench/results/Checkpoints/openvla_flappy_l2_full/wandb", + "host": "rich-fire-blows-fin-03", + "executable": "/root/miniconda3/envs/starvla_rl_games_openvla/bin/python", + "cpu_count": 30, + "cpu_count_logical": 30, + "gpu": "NVIDIA B200", + "gpu_count": 1, + "disk": { + "/": { + "total": "206901264384", + "used": "118122270720" + } + }, + "memory": { + "total": "179351265280" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA B200", + "memoryTotal": "192265846784", + "cudaCores": 18944, + "architecture": "Blackwell", + "uuid": "GPU-2c6fbcfc-60c4-202b-3597-e56a2f0e3579" + } + ], + "cudaVersion": "13.0", + "writerId": "h45zo9uo6uor3rmj7les735282hs81ot" + } + }, + { + "id": "q4wfbo43", + "name": "openvla_flappy_l2_full", + "entity": "saberrr-zju", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-6.8.0-100-generic-x86_64-with-glibc2.39", + "python": "CPython 3.10.20", + "startedAt": "2026-06-02T18:40:27.849795Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "mode=single", + "run_root_dir=/root/latency-sensitive-bench/results/Checkpoints", + "checkpoint.local.keep_last_n=3", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.obs_image_size=[224,224]", + "datasets.vla_data.per_device_batch_size=32", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.video_backend=torchvision_av", + "framework.action_model.action_layout=flappy_categorical_2", + "framework.action_model.loss_type=ce", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.qwenvl.lora.alpha=64", + "framework.qwenvl.lora.bias=none", + "framework.qwenvl.lora.dropout=0.05", + "framework.qwenvl.lora.enabled=false", + "framework.qwenvl.lora.r=32", + "framework.qwenvl.lora.target_modules=all-linear", + "rl_games.env_eval.enabled=false", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=84", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.latency.values=[2]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.mid_train.max_steps_per_episode=2000", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.latencies=[2]", + "rl_games.env_eval.post_train.max_steps_per_episode=2000", + "rl_games.env_eval.post_train.num_episodes=50", + "rl_games.model_alias=openvla", + "rl_games.task=flappy", + "run_id=openvla_flappy_l2_full", + "seed=42", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.gradient_accumulation_steps=2", + "trainer.gradient_clipping=1.0", + "trainer.learning_rate.action_model=0.0001", + "trainer.learning_rate.base=2e-05", + "trainer.learning_rate.qwen_vl_interface=2e-05", + "trainer.logging_frequency=10", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.max_grad_norm=1.0", + "trainer.max_train_steps=2000", + "trainer.num_warmup_steps=100", + "trainer.save_format=pt", + "trainer.save_interval=100", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.weight_decay=0.0", + "wandb_entity=saberrr-zju", + "wandb_project=starVLA_rl_games", + "datasets.vla_data.data_root_dir=/root/latency-sensitive-bench/playground/Datasets/rl_games", + "datasets.vla_data.data_mix=flappy_fixed_l2_fs4", + "datasets.vla_data.eval_data_mix=flappy_fixed_l2_fs4__val", + "framework.qwenvl.base_vlm=/root/latency-sensitive-bench/playground/Pretrained_models/Qwen3-VL-4B-Instruct-Action", + "rl_games.env_eval.latency.prompt_map_path=/root/latency-sensitive-bench/playground/Datasets/rl_games/flappy_fixed_l2_fs4/latency_prompt_map.json", + "trainer.is_resume=false" + ], + "program": "/root/latency-sensitive-bench/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "git@github.com:talha1503/starVLA.git", + "commit": "538fd523fe4fa4e7abce3201611919314bc6b664" + }, + "email": "lxy1501454486@gmail.com", + "root": "/root/latency-sensitive-bench/results/Checkpoints/openvla_flappy_l2_full/wandb", + "host": "rich-fire-blows-fin-03", + "executable": "/root/miniconda3/envs/starvla_rl_games_openvla/bin/python", + "cpu_count": 30, + "cpu_count_logical": 30, + "gpu": "NVIDIA B200", + "gpu_count": 1, + "disk": { + "/": { + "total": "206901264384", + "used": "129243275264" + } + }, + "memory": { + "total": "179351265280" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA B200", + "memoryTotal": "192265846784", + "cudaCores": 18944, + "architecture": "Blackwell", + "uuid": "GPU-2c6fbcfc-60c4-202b-3597-e56a2f0e3579" + } + ], + "cudaVersion": "13.0", + "writerId": "7ryj4jsqckxavvghv8l6xv5w51v7l25u" + } + }, + { + "id": "ne5bp026", + "name": "openvla_flappy_l2_lora", + "entity": "saberrr-zju", + "project": "starVLA_rl_games", + "metadata": { + "os": "Linux-5.15.0-177-generic-x86_64-with-glibc2.35", + "python": "CPython 3.10.20", + "startedAt": "2026-06-02T17:53:22.015345Z", + "args": [ + "--config-name", + "train", + "model=openvla", + "env=flappy", + "mode=single", + "run_root_dir=/mnt/data/latency_data/results/Checkpoints", + "checkpoint.local.keep_last_n=3", + "checkpoint.sync.enabled=false", + "checkpoint.sync.keep_last_n=0", + "datasets.vla_data.action_type=discrete", + "datasets.vla_data.load_all_data_for_training=true", + "datasets.vla_data.obs_image_size=[224,224]", + "datasets.vla_data.per_device_batch_size=16", + "datasets.vla_data.sequential_step_sampling=false", + "datasets.vla_data.video_backend=torchvision_av", + "framework.action_model.action_layout=flappy_categorical_2", + "framework.action_model.loss_type=ce", + "framework.name=QwenOFT", + "framework.qwenvl.attn_implementation=flash_attention_2", + "framework.qwenvl.enable_gradient_checkpointing=true", + "framework.qwenvl.lora.alpha=64", + "framework.qwenvl.lora.bias=none", + "framework.qwenvl.lora.dropout=0.05", + "framework.qwenvl.lora.enabled=true", + "framework.qwenvl.lora.r=32", + "framework.qwenvl.lora.target_modules=all-linear", + "rl_games.env_eval.enabled=false", + "rl_games.env_eval.frameskip=1", + "rl_games.env_eval.image_size=84", + "rl_games.env_eval.latency.mode=single", + "rl_games.env_eval.latency.values=[2]", + "rl_games.env_eval.mid_train.enabled=false", + "rl_games.env_eval.mid_train.interval_steps=100", + "rl_games.env_eval.mid_train.latencies=[2]", + "rl_games.env_eval.mid_train.max_steps_per_episode=2000", + "rl_games.env_eval.mid_train.num_episodes=5", + "rl_games.env_eval.post_train.enabled=false", + "rl_games.env_eval.post_train.latencies=[2]", + "rl_games.env_eval.post_train.max_steps_per_episode=2000", + "rl_games.env_eval.post_train.num_episodes=50", + "rl_games.model_alias=openvla", + "rl_games.task=flappy", + "run_id=openvla_flappy_l2_lora", + "seed=42", + "trainer.eval_interval=100", + "trainer.eval_num_batches=20", + "trainer.gradient_accumulation_steps=4", + "trainer.gradient_clipping=1.0", + "trainer.learning_rate.action_model=0.0001", + "trainer.learning_rate.base=0.0002", + "trainer.learning_rate.qwen_vl_interface=0.0002", + "trainer.logging_frequency=10", + "trainer.loss_scale.vla=1.0", + "trainer.loss_scale.vlm=0.1", + "trainer.lr_scheduler_type=cosine_with_min_lr", + "trainer.max_grad_norm=1.0", + "trainer.max_train_steps=2000", + "trainer.num_warmup_steps=100", + "trainer.save_format=pt", + "trainer.save_interval=100", + "trainer.scheduler_specific_kwargs.min_lr=1e-06", + "trainer.weight_decay=0.0", + "wandb_entity=saberrr-zju", + "wandb_project=starVLA_rl_games", + "datasets.vla_data.data_root_dir=/mnt/data/latency_data/playground/Datasets/rl_games", + "datasets.vla_data.data_mix=flappy_fixed_l2_fs4", + "datasets.vla_data.eval_data_mix=flappy_fixed_l2_fs4__val", + "framework.qwenvl.base_vlm=/mnt/data/latency_data/playground/Pretrained_models/Qwen3-VL-4B-Instruct-Action", + "rl_games.env_eval.latency.prompt_map_path=/mnt/data/latency_data/playground/Datasets/rl_games/flappy_fixed_l2_fs4/latency_prompt_map.json", + "trainer.is_resume=false" + ], + "program": "/home/lixinyuan/latency-sensitive-bench/starVLA/starVLA/training/train_starvla_hydra.py", + "codePath": "starVLA/training/train_starvla_hydra.py", + "codePathLocal": "starVLA/training/train_starvla_hydra.py", + "git": { + "remote": "git@github.com:talha1503/starVLA.git", + "commit": "9aa8c96e8044e6dc48063a9f7e0b13df513ebf62" + }, + "email": "lxy1501454486@gmail.com", + "root": "/mnt/data/latency_data/results/Checkpoints/openvla_flappy_l2_lora/wandb", + "host": "vm10-50-0-235", + "executable": "/home/lixinyuan/miniconda3/envs/starvla_rl_games_openvla/bin/python", + "cpu_count": 14, + "cpu_count_logical": 28, + "gpu": "NVIDIA A100-SXM4-80GB", + "gpu_count": 2, + "disk": { + "/": { + "total": "528237408256", + "used": "366023327744" + } + }, + "memory": { + "total": "235894132736" + }, + "gpu_nvidia": [ + { + "name": "NVIDIA A100-SXM4-80GB", + "memoryTotal": "85899345920", + "cudaCores": 6912, + "architecture": "Ampere", + "uuid": "GPU-46249476-64dd-c979-d67d-23d63248a5fc" + }, + { + "name": "NVIDIA A100-SXM4-80GB", + "memoryTotal": "85899345920", + "cudaCores": 6912, + "architecture": "Ampere", + "uuid": "GPU-c521270c-6815-298c-6ef7-136d618abf34" + } + ], + "cudaVersion": "12.6", + "writerId": "qnz6xi3aa5hs56av3xa7m0t8lewl03r1" + } + } + ] + }, + { + "entity": "talha1503", + "project": "latency-sensitive-bench", + "errors": { + "data": { + "project": null + } + } + } +] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/REPORT.md b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/REPORT.md new file mode 100644 index 0000000000000000000000000000000000000000..a159015e14b8439158a69f5b8dd30446ca529955 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/REPORT.md @@ -0,0 +1,98 @@ +# 按实验用途归档与整仓清理 + +完成批次 0–4。组织 API 复核：**models 85 → 52，datasets 98 → 53**。 +实际整仓删除 **34 个 models、46 个 datasets**；新建 `latency-information-models/data` 两个仓库。 +净减少 **33 个 models、45 个 datasets**。权重数、目录数不计作仓库数。 +API 报告的组织存储从 **6.234340 TB** 变为 +**6.149843 TB**，观察到减少 **84.497 GB**。 +统计采用十进制单位。服务端对象回收可能异步完成；各源仓库的历史随整仓删除一并清理。 +固定引用所需的目标历史继续保留。其他仓库存储变化见 `repository_and_storage_changes.json`。 + +| 批次 | 用途 | 实际整仓删除 models | 实际整仓删除 datasets | 净减少 models / datasets | +|---|---|---:|---:|---:| +| 0 | 已归档 WanOFT L1、Flappy context5 L0/L1 的重复源 | 1 | 2 | 1 / 2 | +| 1 | 正式固定延迟迁移与历史固定延迟训练 | 18 | 38 | 18 / 38 | +| 2 | Ghost-trail，追加 memory 归档 | 10 | 2 | 10 / 2 | +| 3 | 旧 LoRA/full，追加微调范围归档 | 2 | 2 | 2 / 2 | +| 4 | Inverted Pendulum prompt 试验，新建两个归档仓库 | 3 | 2 | 2 / 1 | +| 合计 | | **34** | **46** | **33 / 45** | + +## 归档与来源链 + +- `latency-transfer-models/data`：Flappy 7 个 OpenVLA run、Demon 5 个正式 run，以及 16 个历史模型源。 + 正式 Flappy checkpoint 为 step 5000，Demon 为 step 7000。保留之前的 WanOFT L0–L4 与 H 节 probes。 + 历史模型位于 `historical/<原仓库>/`；训练数据位于 `training/<原仓库>/`。 + 正式 Flappy 为 30 FPS、L0–L4 raw frames。Demon 训练配置的 L0/2/4/6/8 decision intervals + 对应 0/8/16/24/32 raw frames，60 FPS、frameskip 4。历史数据保留自己的 episode 数与协议。 +- `memory-models/data`：10 个 Ghost 模型源和 2 个 Ghost-15 独立数据源。 + Ghost-7 no-exp 与 exp3 的权重不同，分别保留。其余 Ghost 训练数据按原配置引用仍保留的 `memory-rollouts` 固定 revision。 +- `vla-finetuning-scope-models/data`：旧 LoRA/full 的模型、adapter/action head、`flappy_fixed_l2_fs4` 与评测 JSON。 + 与正式 7 个 scope run 分开。正式 scope 的共享 Flappy L2 数据改为 `latency-transfer-data` 固定 revision。 +- `latency-information-models/data`：三个 Inverted Pendulum run 与两套原始数据。 + 两个 `raw` prompt run、一个 `latency_neutral` run分别保留。两个源中的 L2/L4 文件身份相同；L0 文件不同。 + +`experiment_repository_index.json` 记录原仓库和 revision、训练配置、数据引用、权重身份、teacher 元数据、schema/行数和论文位置。 +`training_bindings.json` 与 `wandb_provenance.json` 保存已记录的日期、W&B run、代码 commit。 +训练代码快照存于 `latency-transfer-data/code/`。没有 W&B 或精确数据来源的资产保留原始缺失状态。 +原配置与原引用文档作为 provenance 保留；当前消费者使用新归档的固定 revision。 + +## 不能据此推断的内容 + +- Flappy cumulative curriculum 配置声明 L0–L5、每个 latency 40 episodes，但存储统计是 200 trajectories、341369 transitions。 + exclusive 声明 L0–L4，存储统计同样是 200 trajectories、341369 transitions。 + 原本的本地 curriculum 数据未发布。不能把配置中的 240 episodes 当作已验证的数据组成，也不能拿固定延迟数据补造配对。 +- Inverted Pendulum 的原论文表没有 checkpoint 标识。归档三个候选模型不证明它们就是表中的评测权重。 + 论文 120 条原始 episode returns 保持不变。 +- 旧 LoRA/full 的 HF 评测各有 50 episodes，Full/LoRA 均值为 18.5180/3.8620。 + 论文旧图 CSV 各有 100 episodes，均值为 22.3080/3.4820。两套记录分别保留。 + 论文旧图的逐 episode 原文件与精确权重绑定未恢复，不能声称 HF 50 局就是图中的 100 局。 +- Ghost 模型的实验类型已确认，但归档全部版本不等于确定每个版本的表格用途。 + 表格仍以已有 accepted source manifest 为准。 +- `openvla_demon_attack_fix_latency_6_small` 原仓库没有可用 checkpoint； + `openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2` 仅有元数据。 + 原始状态保留，不补造配置或权重。 +- 一些早期模型只记录本地训练目录。能够匹配原目录名的配对已记录证据，未恢复的原始训练数据明确留空。 + 所有不同权重仍分别保留。 + +## 验证结果 + +- 735 个迁移源文件的对象身份匹配。训练状态中的推理权重规范为 `steps__model.safetensors`；权重字节未修改。 +- 55 个 checkpoint/adapter 文件通过原生加载，归一化统计一致。LoRA 的 action head 同时加载。 + 两份旧 safetensors 使用现有 tied-LM-head 兼容加载逻辑，具体 loader revision 已记录。 + 批次 0 Wan L1 与之前通过原生加载的归档权重身份相同。 +- 135 个不同 Parquet 对象、160 个路径引用通过 schema、总行数和真实样本读取检查。 + 含图像的样本通过解码。检查没有声称逐行解码全部数据。 +- Flappy 700 条和 Demon 500 条论文记录的 60 个汇总单元均值与 source manifest 一致；相关 CSV 数值未改变。 +- 迁移相关测试 8 项、数据转换测试 22 项通过。另有一次较宽测试：98 项通过、4 项现有失败、1 项未运行。 + 四项失败涉及三个 pi0 initializer 断言和一个 runtime-cache 断言，未把它们算作通过。 +- 58 个 shell、3 个 YAML、7 个 Python 消费者文件检查通过。修复四个既有消费者脚本缺失的关闭引号。 +- arXiv 与 ICLR 稿编译通过，`git diff --check` 通过。 + 9 份来源文档已提交并同步 Overleaf：`7d13a2f74fc3933f12916e08f103044af782b624`。既有未提交的五份论文编辑保留。 +- 删除后，从四个模型归档重新下载并原生加载各一份 checkpoint/adapter，从四个数据归档读取各一个 Parquet。 + 全部通过。80 个删除源在最终组织 API 列表中均不存在。 + +## 保留并继续调查的资产 + +已重新固定 **51 个**待调查资产的 revision，详见 `pending_asset_inventory.json`。 + +| 组 | 独立仓库数 | 当前结论 | +|---|---:|---| +| 其余 image-history/fs 数据 | 12 datasets | teacher frame_stack 与导出 image_stack 不是同一参数；不归入论文 APPO context-window 扫描 | +| Deadly jitter | 5 datasets | teacher 与随机延迟参数保留；学生消费者与论文绑定尚未确认 | +| mixed 数据 | 2 datasets | 存在旧跨任务消费者；用途不与固定延迟或 prompt 消融混写 | +| 零延迟候选 | 19 models | 正式基线、重跑、其他任务和未完成产物分别核对 | +| RTX3090 profile 历史组 | 11 models | 保留差结果/失效/未正式发布状态；不并入 formal benchmark | +| 跨任务模型 | 2 models | 保留原跨任务用途，未按 latency transfer 删除 | + +独立 WanOFT `demon_attack_200ep` 模型保留。此次删除同名 **dataset** 不代表删除同名 **model**。 +前一批 factor-sweeps 中 9 个 Deadly Corridor run 的候选权重仍分别保留，本次没有消除其评测权重歧义。 + +## 主要证据 + +- `deletion_receipts.json`：80 个整仓删除回执。 +- `organization_before.json` / `organization_final.json`：数量与存储复核。 +- `deletion_gate.json`：固定源 revision、完整迁移与数值核对。 +- `legacy_lora_evaluation_difference.json`：旧 LoRA/full 的 50 局和 100 局记录差异。 +- `native_loader_verification.json` / `data_readability_verification.json`：全批加载/读取记录。 +- `postcleanup_verification.json`：删除后的真实下载、加载与读取。 +- `credential_boundary_check.json`：HF 客户端使用官方端点；公开文件下载不接收工作区 token；已检查的元数据文本未发现凭据字面量。 diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/credential_boundary_check.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/credential_boundary_check.json new file mode 100644 index 0000000000000000000000000000000000000000..2cae00aa78829f6b55c58b83f9c5d6ecd0a6a806 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/credential_boundary_check.json @@ -0,0 +1,6 @@ +{ + "hf_endpoint": "https://huggingface.co", + "credential_literals_detected": [], + "tracked_env_archives": [], + "downloads": "Public artifact downloads did not receive the workspace token." +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/deletion_gate.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/deletion_gate.json new file mode 100644 index 0000000000000000000000000000000000000000..33ece264eaf2f42b68dd8cb06d75892351ac06ab --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/deletion_gate.json @@ -0,0 +1,664 @@ +{ + "sources": [ + { + "batch": "0", + "type": "model", + "repo": "latency-sensitive-bench/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce", + "source_revision": "8cf964f252337912a8f8e5d30467a33f4e8a49dc", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "0", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_0_200ep_context5", + "source_revision": "f5f82fb2c760bef7ddb8882dbba18075fa1cf2c8", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "0", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_1_200ep_context5", + "source_revision": "8ee61cea50b7c3fcad97f4ab2f41f97044d219a8", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0", + "source_revision": "06f64d1a03905ad0f04a447c57afb970c29a559e", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_6", + "source_revision": "7a2e2f81dda9fc9de068fa15b1eac442ed6fb51f", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2", + "source_revision": "8135db54520527abe4a62f002323a307d1a47c2d", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_exp", + "source_revision": "213bab6c06b13c73fee875412d99e40ea4a4dd70", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_small", + "source_revision": "371074f07545d1d4e5533afc4d457a63d03bf9da", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small_tmp", + "source_revision": "8fba8c350f6f10d70119807a546218cfa82e9150", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0_small", + "source_revision": "05586f38706727534a0f388ae0ece46541776d65", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small", + "source_revision": "7842a04e6ba8ed4dc56ccdae235735a807a77b5d", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2_small", + "source_revision": "87c38afed1a227e92aac9efeb0921f57d556df8a", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_3_small", + "source_revision": "86b4d314c377420fa8eb9ac7aa268695c2ab4f40", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0_full", + "source_revision": "9281fd061894c14d3641ee9724e108cd33baf494", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0", + "source_revision": "9f8c6842151405a82a3e852d2a00e27885db1e93", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_2", + "source_revision": "0a60d6c8a400b48e4d9737b36ef3a44814d52d2a", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_0_small", + "source_revision": "aeefbbd9676d95973aea2a6f63c433807a44d05b", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_2_small", + "source_revision": "bb91ae40a1bb9b9f6a93be031193de603da86ecd", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_6_small", + "source_revision": "9e8e2e6cc073eb68bd59306ab1141972ed1ed819", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_2", + "source_revision": "c0605fa1df28b19249849c3e748bbc445bef6526", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_1", + "source_revision": "7190bf45dfface92ec7cbaa3c774a7e29616db52", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_0", + "source_revision": "2deacb435543c52ae51f3c947bf917fdb0b5493c", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_3", + "source_revision": "709111a3b59e413214e0417bd878291734ec9427", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_4", + "source_revision": "c7f6eba84f512bc22a095971c9b75499e841a026", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_5", + "source_revision": "67876ea9a8d5c687b924199574e298c68bf9b83d", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_7", + "source_revision": "d76a57ef1b84c591c47975347c539da59670f3b4", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_6", + "source_revision": "0766ba4ea9196b861302e99e564b7d2e05453885", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_8", + "source_revision": "e02c6b1db39b37965569d4f559f1d710fb55c58a", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_12", + "source_revision": "ac9d21a59b908e717de2bee2cc749b90495a00b8", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_9", + "source_revision": "7be93d38340141e385913e540b028024d74f1cf8", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_13", + "source_revision": "2a5c5d2aa4730db22fd33f4f00f4d93fc5043465", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_14", + "source_revision": "9867d03ad02a4d025cb6cc88281f95596f83a814", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_10", + "source_revision": "9a3f7f736402b063885068e683860b6746e5337d", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_15", + "source_revision": "21811c6039f0a538f11821730ee6792bd0d6c0b4", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_11", + "source_revision": "78e4e1b6d9bb57c5f30712cd0ed2f3b02877ff56", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_0", + "source_revision": "6ee54a31fe5645bce299adeab17a68ff458bfabe", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_12", + "source_revision": "258c4ae285ad34445578dea4e786c408fcd6be30", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_4", + "source_revision": "4dee6c0fd4abb86507001c4dc7404a5310c2f30a", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_13", + "source_revision": "dbb7678df6396735290099d070dfb5827a4cf4a4", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_2", + "source_revision": "df04b6f30856b2903e65a6a870f7440055f02311", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_14", + "source_revision": "fb78a1bc6280a4d0a6a60e8231d63d848819bd1e", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_5", + "source_revision": "f614a91258d86cb34cd2e3e0e79e2bf69b321e76", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_7", + "source_revision": "d04b9bc42bfbb7da400438a9fb75c49b198f7776", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_1", + "source_revision": "e09b7f2c51726bb169d827fe973b2d429b6b1432", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_10", + "source_revision": "94762ebd79edb55d6e83ba2046729e3e6574d64d", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_15", + "source_revision": "c586ddd84f76b6295de7929b62ad4da3359b0348", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_6", + "source_revision": "47cfced1713ce7d57d4567110269fb5929e8e7a8", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_9", + "source_revision": "3405ac7c12ca266e2d3e83047bbd06e1e5eed220", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_3", + "source_revision": "b8b18213fe10aa0a6be2b7602e23c2ac938b49e0", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_8", + "source_revision": "164f4e56abb6c2637801e779918a99a7d70c9c3a", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_3_small", + "source_revision": "8470ae1f00ca10bb8646ac7cfdccf9682e053eab", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_2_small", + "source_revision": "c9bbd1744020a587b55590c897628510e0741136", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_1_small", + "source_revision": "749a849911b8c35005c26241fb4edbcee1286832", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_0_small", + "source_revision": "7e974f1e8e1ad7ca4ee0ffad0849d886d64128d3", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_4_small", + "source_revision": "8a6a091570530476b313976b6a8d1b94d600f6d9", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "source_revision": "5dfd90884fc4e3817bc51f574d6fc9cea2d7948b", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3", + "source_revision": "f8d64aff21628adbdd0d1aaddd1bc7067bc2a23c", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "9915dcb0dc644aa2ed14172f298386f53b03c6f6", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "source_revision": "cd810024c06f961c95449785f19ab73ccfd20de1", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2", + "source_revision": "73ea97c064d206c3a0dcf6ffdbf2ecfe97bc2c3c", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2", + "source_revision": "c4cbca936a122c2d55b1edee3959656f8174c5b7", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1", + "source_revision": "004ec237af79c10de5c0d4376cf8a13542ac01f8", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1", + "source_revision": "bc733d64e32512a8fc76f9be022eb5b0d7c0e2cb", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7", + "source_revision": "94014d226e26fcf46f117f6bea42a59b0be2d753", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2", + "source_revision": "ac5ff1109c46295299d74b5e30a56f1e27e930a6", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "f90d1dac8b31a18c0496884729c744d7a4c877ae", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "2", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "source_revision": "862b6cad9b2c87dc129d5cf45a9b23b18450c4ef", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "3", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_l2_full", + "source_revision": "2e8e58c6af1ee6f8a3701388b2552a4fa65a041c", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "3", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_l2_lora", + "source_revision": "427f4c0582d6eb9419fa39f5942db1d59585993c", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "3", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fixed_l2_fs4", + "source_revision": "1bdd2ea6ea7db07139e4694f445c12198399c2b8", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "3", + "type": "dataset", + "repo": "latency-sensitive-bench/lora_comparison_flappy_l2_eval", + "source_revision": "97db426f51a9427fbe1e8c005b33dbe7aeac3899", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2", + "source_revision": "177c11a86ee64432264ae8f1e577ecfd6c767918", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3", + "source_revision": "1967aac3fadf905a7f693e388cbf2aebdd40e5de", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "source_revision": "bd6274f20b7ff2f36c0b758058ef8ce2e996d7f6", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "4", + "type": "dataset", + "repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + }, + { + "batch": "4", + "type": "dataset", + "repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_head_unchanged": true, + "retention": "All original inference weights and required metadata copied, or verified duplicate. Missing original data/metadata remain explicit." + } + ], + "inference_weights_native_verified": 55, + "unique_parquet_objects_read": 135, + "copied_source_files_verified": 695, + "paper_numerical_consistency": [ + { + "experiment": "flappy", + "episodes": 700, + "cells": 35, + "all_means_match": true, + "csv_unchanged": true + }, + { + "experiment": "demon_attack", + "episodes": 500, + "cells": 25, + "all_means_match": true, + "csv_unchanged": true + } + ], + "ready": true +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/deletion_receipts.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/deletion_receipts.json new file mode 100644 index 0000000000000000000000000000000000000000..3db299776367996fc6f7e9a6ca07b347823acc41 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/deletion_receipts.json @@ -0,0 +1,562 @@ +[ + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0", + "source_revision": "06f64d1a03905ad0f04a447c57afb970c29a559e", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_6", + "source_revision": "7a2e2f81dda9fc9de068fa15b1eac442ed6fb51f", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2", + "source_revision": "8135db54520527abe4a62f002323a307d1a47c2d", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_exp", + "source_revision": "213bab6c06b13c73fee875412d99e40ea4a4dd70", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_4_small", + "source_revision": "371074f07545d1d4e5533afc4d457a63d03bf9da", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small_tmp", + "source_revision": "8fba8c350f6f10d70119807a546218cfa82e9150", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_0_small", + "source_revision": "05586f38706727534a0f388ae0ece46541776d65", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_1_small", + "source_revision": "7842a04e6ba8ed4dc56ccdae235735a807a77b5d", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_2_small", + "source_revision": "87c38afed1a227e92aac9efeb0921f57d556df8a", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_fix_latency_3_small", + "source_revision": "86b4d314c377420fa8eb9ac7aa268695c2ab4f40", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0_full", + "source_revision": "9281fd061894c14d3641ee9724e108cd33baf494", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_0", + "source_revision": "9f8c6842151405a82a3e852d2a00e27885db1e93", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_deadly_corridor_fix_latency_2", + "source_revision": "0a60d6c8a400b48e4d9737b36ef3a44814d52d2a", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_0_small", + "source_revision": "aeefbbd9676d95973aea2a6f63c433807a44d05b", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_2_small", + "source_revision": "bb91ae40a1bb9b9f6a93be031193de603da86ecd", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_fix_latency_6_small", + "source_revision": "9e8e2e6cc073eb68bd59306ab1141972ed1ed819", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_200ep", + "source_revision": "7918bf5c0aeb830a01352f5378acc6a4d4cf601c", + "deleted": true + }, + { + "batch": "1", + "type": "model", + "repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_2", + "source_revision": "c0605fa1df28b19249849c3e748bbc445bef6526", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_1", + "source_revision": "7190bf45dfface92ec7cbaa3c774a7e29616db52", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_0", + "source_revision": "2deacb435543c52ae51f3c947bf917fdb0b5493c", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_3", + "source_revision": "709111a3b59e413214e0417bd878291734ec9427", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_4", + "source_revision": "c7f6eba84f512bc22a095971c9b75499e841a026", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_5", + "source_revision": "67876ea9a8d5c687b924199574e298c68bf9b83d", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_7", + "source_revision": "d76a57ef1b84c591c47975347c539da59670f3b4", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_6", + "source_revision": "0766ba4ea9196b861302e99e564b7d2e05453885", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_8", + "source_revision": "e02c6b1db39b37965569d4f559f1d710fb55c58a", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_12", + "source_revision": "ac9d21a59b908e717de2bee2cc749b90495a00b8", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_9", + "source_revision": "7be93d38340141e385913e540b028024d74f1cf8", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_13", + "source_revision": "2a5c5d2aa4730db22fd33f4f00f4d93fc5043465", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_14", + "source_revision": "9867d03ad02a4d025cb6cc88281f95596f83a814", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_10", + "source_revision": "9a3f7f736402b063885068e683860b6746e5337d", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_15", + "source_revision": "21811c6039f0a538f11821730ee6792bd0d6c0b4", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fix_latency_11", + "source_revision": "78e4e1b6d9bb57c5f30712cd0ed2f3b02877ff56", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_0", + "source_revision": "6ee54a31fe5645bce299adeab17a68ff458bfabe", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_12", + "source_revision": "258c4ae285ad34445578dea4e786c408fcd6be30", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_4", + "source_revision": "4dee6c0fd4abb86507001c4dc7404a5310c2f30a", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_13", + "source_revision": "dbb7678df6396735290099d070dfb5827a4cf4a4", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_2", + "source_revision": "df04b6f30856b2903e65a6a870f7440055f02311", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_14", + "source_revision": "fb78a1bc6280a4d0a6a60e8231d63d848819bd1e", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_5", + "source_revision": "f614a91258d86cb34cd2e3e0e79e2bf69b321e76", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_7", + "source_revision": "d04b9bc42bfbb7da400438a9fb75c49b198f7776", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_1", + "source_revision": "e09b7f2c51726bb169d827fe973b2d429b6b1432", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_10", + "source_revision": "94762ebd79edb55d6e83ba2046729e3e6574d64d", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_15", + "source_revision": "c586ddd84f76b6295de7929b62ad4da3359b0348", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_6", + "source_revision": "47cfced1713ce7d57d4567110269fb5929e8e7a8", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_9", + "source_revision": "3405ac7c12ca266e2d3e83047bbd06e1e5eed220", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_3", + "source_revision": "b8b18213fe10aa0a6be2b7602e23c2ac938b49e0", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/deadly_corridor_fix_latency_8", + "source_revision": "164f4e56abb6c2637801e779918a99a7d70c9c3a", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_3_small", + "source_revision": "8470ae1f00ca10bb8646ac7cfdccf9682e053eab", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_2_small", + "source_revision": "c9bbd1744020a587b55590c897628510e0741136", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_1_small", + "source_revision": "749a849911b8c35005c26241fb4edbcee1286832", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_0_small", + "source_revision": "7e974f1e8e1ad7ca4ee0ffad0849d886d64128d3", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_4_small", + "source_revision": "8a6a091570530476b313976b6a8d1b94d600f6d9", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_200ep", + "source_revision": "5df5688a961dc2c91090155c719bff4a5b248235", + "deleted": true + }, + { + "batch": "1", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_200ep", + "source_revision": "0481972d94f6409a392c5b035f9a7f5c2681a96d", + "deleted": true + }, + { + "batch": "0", + "type": "model", + "repo": "latency-sensitive-bench/wan_oft_flappy_fix_latency_1_context5_standard_sft_2000_effbs128_224_currentce", + "source_revision": "8cf964f252337912a8f8e5d30467a33f4e8a49dc", + "deleted": true + }, + { + "batch": "0", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_0_200ep_context5", + "source_revision": "f5f82fb2c760bef7ddb8882dbba18075fa1cf2c8", + "deleted": true + }, + { + "batch": "0", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fix_latency_1_200ep_context5", + "source_revision": "8ee61cea50b7c3fcad97f4ab2f41f97044d219a8", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "source_revision": "5dfd90884fc4e3817bc51f574d6fc9cea2d7948b", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2_exp3", + "source_revision": "f8d64aff21628adbdd0d1aaddd1bc7067bc2a23c", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "9915dcb0dc644aa2ed14172f298386f53b03c6f6", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "source_revision": "cd810024c06f961c95449785f19ab73ccfd20de1", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_exp2", + "source_revision": "73ea97c064d206c3a0dcf6ffdbf2ecfe97bc2c3c", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_exp2", + "source_revision": "c4cbca936a122c2d55b1edee3959656f8174c5b7", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost15_seed1", + "source_revision": "004ec237af79c10de5c0d4376cf8a13542ac01f8", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_seed1", + "source_revision": "bc733d64e32512a8fc76f9be022eb5b0d7c0e2cb", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7", + "source_revision": "94014d226e26fcf46f117f6bea42a59b0be2d753", + "deleted": true + }, + { + "batch": "2", + "type": "model", + "repo": "latency-sensitive-bench/openvla_bridge_demon_attack_fixed_latency_6_200ep_7k2steps_ghost15_2", + "source_revision": "ac5ff1109c46295299d74b5e30a56f1e27e930a6", + "deleted": true + }, + { + "batch": "2", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "f90d1dac8b31a18c0496884729c744d7a4c877ae", + "deleted": true + }, + { + "batch": "2", + "type": "dataset", + "repo": "latency-sensitive-bench/demon_attack_fixed_latency_6_200ep_7k2steps_ghost15", + "source_revision": "862b6cad9b2c87dc129d5cf45a9b23b18450c4ef", + "deleted": true + }, + { + "batch": "3", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_l2_full", + "source_revision": "2e8e58c6af1ee6f8a3701388b2552a4fa65a041c", + "deleted": true + }, + { + "batch": "3", + "type": "model", + "repo": "latency-sensitive-bench/openvla_flappy_l2_lora", + "source_revision": "427f4c0582d6eb9419fa39f5942db1d59585993c", + "deleted": true + }, + { + "batch": "3", + "type": "dataset", + "repo": "latency-sensitive-bench/flappy_fixed_l2_fs4", + "source_revision": "1bdd2ea6ea7db07139e4694f445c12198399c2b8", + "deleted": true + }, + { + "batch": "3", + "type": "dataset", + "repo": "latency-sensitive-bench/lora_comparison_flappy_l2_eval", + "source_revision": "97db426f51a9427fbe1e8c005b33dbe7aeac3899", + "deleted": true + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp2", + "source_revision": "177c11a86ee64432264ae8f1e577ecfd6c767918", + "deleted": true + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_exp3", + "source_revision": "1967aac3fadf905a7f693e388cbf2aebdd40e5de", + "deleted": true + }, + { + "batch": "4", + "type": "model", + "repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "source_revision": "bd6274f20b7ff2f36c0b758058ef8ce2e996d7f6", + "deleted": true + }, + { + "batch": "4", + "type": "dataset", + "repo": "latency-sensitive-bench/inverted_pendulum_200ep", + "source_revision": "5dcc48f5f713386b9ee2c2339f1c364b76246956", + "deleted": true + }, + { + "batch": "4", + "type": "dataset", + "repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "deleted": true + } +] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/duplicate_wan_l1_native_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/duplicate_wan_l1_native_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..4b4f5e2518fb794e1cce9cbab02d8376e8aed0a1 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/duplicate_wan_l1_native_verification.json @@ -0,0 +1,16 @@ +{ + "repo": "latency-sensitive-bench/latency-transfer-models", + "type": "model", + "revision": "c2d33435cd46b44c1d1c53973439978364b0597d", + "path": "wan_oft_flappy_fix_latency_1_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "identity": "ca01daf3e577fedcc238882e531c748f6a5c178934ca7cc7209da18eb24ff03f", + "code_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "framework": "wan", + "checkpoint": "/tmp/latency-transfer-validation-20261007/native-one/wan_oft_flappy_fix_latency_1_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "backbone": "/tmp/latency-transfer-validation-20261007/hf/models--Wan-AI--Wan2.2-TI2V-5B-Diffusers/snapshots/b8fff7315c768468a5333511427288870b2e9635", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 1282, + "loader": "baseframework.from_pretrained", + "validation_backbone_revision": "b8fff7315c768468a5333511427288870b2e9635" +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/final_copy_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/final_copy_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..5f7987b2f5b41c2880b1c01ab2eee67565bd79e8 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/final_copy_verification.json @@ -0,0 +1,6 @@ +{ + "all_target_files_match_sources_after_deletion": true, + "copied_files": 695, + "training_version_files": 40, + "total": 735 +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/legacy_lora_evaluation_difference.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/legacy_lora_evaluation_difference.json new file mode 100644 index 0000000000000000000000000000000000000000..f04d2744f32988794b297483d0eda70270539487 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/legacy_lora_evaluation_difference.json @@ -0,0 +1,22 @@ +[ + { + "method": "full", + "hf_episode_count": 50, + "hf_mean_return": 18.518000243455173, + "paper_episode_count": 100, + "paper_mean_return": 22.308000287115576, + "same_evaluation_record": false, + "hf_timestamp": "2026-06-02T23:20:12.557796+00:00", + "note": "The HF 50-episode record and paper 100-episode aggregate are retained separately. The paper local per-episode source files were not recovered." + }, + { + "method": "lora", + "hf_episode_count": 50, + "hf_mean_return": 3.8620000724494457, + "paper_episode_count": 100, + "paper_mean_return": 3.4820000667870046, + "same_evaluation_record": false, + "hf_timestamp": "2026-06-02T22:57:10.859128+00:00", + "note": "The HF 50-episode record and paper 100-episode aggregate are retained separately. The paper local per-episode source files were not recovered." + } +] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/local_consumer_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/local_consumer_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..0074036c6f0ac47e1e90359be8854959c243b54c --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/local_consumer_verification.json @@ -0,0 +1,5 @@ +{ + "shell": 58, + "yaml": 3, + "python": 7 +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/paper_numerical_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/paper_numerical_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..6a9a2846497bcd90a6b1465471e3c20f99899144 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/paper_numerical_verification.json @@ -0,0 +1,16 @@ +[ + { + "experiment": "flappy", + "episodes": 700, + "cells": 35, + "all_means_match": true, + "csv_unchanged": true + }, + { + "experiment": "demon_attack", + "episodes": 500, + "cells": 25, + "all_means_match": true, + "csv_unchanged": true + } +] diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/paper_publication_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/paper_publication_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..ae02780d9a7584ea44eba996f9ce1d0963277be8 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/paper_publication_verification.json @@ -0,0 +1,16 @@ +{ + "commit": "7d13a2f74fc3933f12916e08f103044af782b624", + "overleaf_revision": "02c620eb46df333039d3acb956bd385964d28253", + "documentation_files_committed": 9, + "unrelated_local_changes_preserved": [ + " M exps/memory/compare_table/wanoft/README.md", + " M exps/memory/compare_table/wanoft/build_data.py", + " M exps/memory/compare_table/wanoft/data.csv", + " M exps/memory/compare_table/wanoft/source_manifest.csv", + " M figures/fig7_memory_interfaces/README.md" + ], + "arxiv_compile_passed": true, + "iclr_compile_passed": true, + "evaluation_difference_followup_commit": "02c620eb46df333039d3acb956bd385964d28253", + "legacy_plot_numeric_inputs_unchanged": true +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/postcleanup_verification.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/postcleanup_verification.json new file mode 100644 index 0000000000000000000000000000000000000000..c73da8119f7ef663f98a2de3bdbcdc7ae6bb5d8f --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/postcleanup_verification.json @@ -0,0 +1,348 @@ +{ + "models": [ + { + "source_repo": "latency-sensitive-bench/openvla_demon_attack_200ep", + "source_revision": "04fece441aa5b6f5a6a015264a6aa5df3dc4c29a", + "repo": "latency-sensitive-bench/latency-transfer-models", + "revision": "fe956a175efd64a52cc41423ad09f53a8b9d14c0", + "path": "demon_attack_fix_latency_0_200ep/checkpoints/steps_7000_model.safetensors", + "identity": "25a15d0460a63a7ca036e9300d39a4f57b8f9e656c6ff42941348b6a570bf767", + "runpath": "demon_attack_fix_latency_0_200ep", + "configpath": "demon_attack_fix_latency_0_200ep/config.yaml", + "statpath": "demon_attack_fix_latency_0_200ep/dataset_statistics.json", + "run_id": "demon_attack_fix_latency_0_200ep", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "dongqianyu99-zhejiang-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "1a93a1ae6772e74087a836b7b2246d340b948bd2" + ], + "loader_revision": "1a93a1ae6772e74087a836b7b2246d340b948bd2", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/checkpoints/steps_7000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/1a93a1ae6772e74087a836b7b2246d340b948bd2", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_flappy_l2_lora", + "source_revision": "427f4c0582d6eb9419fa39f5942db1d59585993c", + "repo": "latency-sensitive-bench/vla-finetuning-scope-models", + "revision": "c56fff9239108f256ac6701ca31042eaef2074c3", + "path": "historical/openvla_flappy_l2_lora/final_model/adapter_model.safetensors", + "identity": "551bd2362cc9548b9d97f3a4f77fcfbe2b28f2725b2897ab42b996485962b8bb", + "runpath": "historical/openvla_flappy_l2_lora", + "configpath": "historical/openvla_flappy_l2_lora/config.yaml", + "statpath": "historical/openvla_flappy_l2_lora/dataset_statistics.json", + "run_id": "openvla_flappy_l2_lora", + "kind": "lora", + "native_ready": true, + "base_model": "action", + "framework": "QwenOFT", + "wandb_entity": "saberrr-zju", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [ + "9aa8c96e8044e6dc48063a9f7e0b13df513ebf62" + ], + "loader_revision": "9aa8c96e8044e6dc48063a9f7e0b13df513ebf62", + "loader_is_exact_training_code": true, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-0/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-0/run/final_model", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/9aa8c96e8044e6dc48063a9f7e0b13df513ebf62", + "backbone": "/tmp/experiment-repo-validation-20261007/action-base-0/models--StarVLA--Qwen3-VL-4B-Instruct-Action/snapshots/c41500cf1d287cd79e9f6602f419937b05048bd8", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-0/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 1442, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "source_revision": "5dfd90884fc4e3817bc51f574d6fc9cea2d7948b", + "repo": "latency-sensitive-bench/memory-models", + "revision": "3883a9a77b2768f921c413d7e120a35229c13d77", + "path": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3/checkpoints/steps_4000_model.safetensors", + "identity": "553d47011aa16cef4609610ea44342630fb4b0b7aabc65da5fe58d133915f40a", + "runpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "configpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3/config.yaml", + "statpath": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3/dataset_statistics.json", + "run_id": "openvla_bridge_flappy_fixed_latency_3_200ep_7k2steps_ghost7_exp3", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "zihanwang-ai-northwestern-university", + "wandb_project": "starVLA_rl_games", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_4000_model.safetensors", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + }, + { + "source_repo": "latency-sensitive-bench/openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "source_revision": "bd6274f20b7ff2f36c0b758058ef8ce2e996d7f6", + "repo": "latency-sensitive-bench/latency-information-models", + "revision": "7f7ff9fd548e66ce31680cc8a05b7386ecfca0f4", + "path": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2/checkpoints/steps_5000_pytorch_model.pt", + "identity": "d275016d512fe0fe071d5c2c9f7362fba939c89cb357e461d8b59e96c31ba62f", + "runpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "configpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2/config.yaml", + "statpath": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2/dataset_statistics.json", + "run_id": "openvla_inverted_pendulum_mixed_latency_024_og_l0_native_no_latency_prompt_exp2", + "kind": "full", + "native_ready": true, + "base_model": "qwen", + "framework": "QwenOFT", + "wandb_entity": "talha1503", + "wandb_project": "latency-sensitive-bench", + "recorded_training_commits": [], + "loader_revision": "cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "loader_is_exact_training_code": false, + "run_dir": "/tmp/experiment-repo-validation-20261007/worker-1/run", + "local_checkpoint": "/tmp/experiment-repo-validation-20261007/worker-1/run/checkpoints/steps_5000_pytorch_model.pt", + "code_dir": "/tmp/experiment-repo-validation-20261007/code/cec6c0eeb44e9a485820591ba05001a14f58cfcb", + "backbone": "/tmp/experiment-repo-validation-20261007/base/models--Qwen--Qwen3-VL-4B-Instruct/snapshots/ebb281ec70b05090aa6165b016eac8ec08e71b17", + "result_path": "/tmp/experiment-repo-validation-20261007/worker-1/result.json", + "native_loaded": true, + "normalization_statistics_equal": true, + "state_dict_keys": 730, + "validation_attention_backend": "sdpa", + "backbone_override_only": true, + "loader": "baseframework.from_pretrained" + } + ], + "datasets": [ + { + "source_repo": "latency-sensitive-bench/inverted_pendulum_200ep_2", + "source_revision": "eebf63d9082f2f2f3a3672cca64f7b7776750b54", + "source_path": "inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "target_repo": "latency-sensitive-bench/latency-information-data", + "type": "dataset", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "identity": "605cd50b8ad5f520756a9190eb7c30dcd58c4d35330ac388e1823f8d09158b32", + "size": 304119714, + "revision": "70708248fc1c22a181315acf652f8730e4e00a05", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep_2/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + }, + { + "repo": "latency-sensitive-bench/latency-information-data", + "path": "training/inverted_pendulum_200ep/inverted_pendulum_fixed_latency_2_1000ep_7k2steps/train-00000-of-00002.parquet", + "revision": "c48e2bbe4a9a3a1adf910911822268aa5d4c4def" + } + ], + "parquet_rows": 49147, + "row_groups": 4, + "columns": [ + "image", + "env_name", + "episode_idx", + "decision_step", + "split", + "seed", + "prompt", + "latency_raw_frames", + "latency_ms", + "raw_reward", + "action_id", + "action", + "action_text", + "state" + ], + "decoded_images": 1, + "first_row_scalars": { + "env_name": "inverted_pendulum", + "episode_idx": 0, + "decision_step": 0, + "split": "train", + "seed": 25, + "latency_raw_frames": 2, + "latency_ms": 80.0, + "raw_reward": 1.0, + "action_id": 2, + "action_text": "zero_force" + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fixed_l2_fs4", + "source_revision": "1bdd2ea6ea7db07139e4694f445c12198399c2b8", + "source_path": "train.parquet", + "target_repo": "latency-sensitive-bench/vla-finetuning-scope-data", + "type": "dataset", + "path": "training/flappy_fixed_l2_fs4/train.parquet", + "identity": "00a0b4f862e5a1e6afda14bcb6106c7a4ebe19b09ef8f131c02554bb2e63de84", + "size": 11126190209, + "revision": "1c513eb272989e50686ef7e71e6981fdd98ada91", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/vla-finetuning-scope-data", + "path": "training/flappy_fixed_l2_fs4/train.parquet", + "revision": "1354ac73e35d37849be824ccf2dc106796f0c3d0" + } + ], + "parquet_rows": 948645, + "row_groups": 3706, + "columns": [ + "action", + "action_id", + "action_text", + "deterministic", + "done", + "episode_idx", + "image_stack", + "latency_ms", + "latency_raw_frames", + "prompt", + "reward", + "seed", + "split", + "t" + ], + "decoded_images": 4, + "first_row_scalars": { + "action_id": 0, + "action_text": "noop", + "deterministic": true, + "done": false, + "episode_idx": 0, + "latency_ms": 66.66666666666667, + "latency_raw_frames": 2, + "reward": 0.20000000298023224, + "seed": 1, + "split": "train", + "t": 0 + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/deadly_corridor_fix_latency_2", + "source_revision": "df04b6f30856b2903e65a6a870f7440055f02311", + "source_path": "data/train-00000-of-00001.parquet", + "target_repo": "latency-sensitive-bench/latency-transfer-data", + "type": "dataset", + "path": "training/deadly_corridor_fix_latency_2/data/train-00000-of-00001.parquet", + "identity": "d7f14fe96cc13b1cad520b3e9d1c51ae675773a66183ada6bae5bca965e7169f", + "size": 38515375, + "revision": "edc8778f922e2640b3350e8d25374d8d5464c18d", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/latency-transfer-data", + "path": "training/deadly_corridor_fix_latency_2/data/train-00000-of-00001.parquet", + "revision": "27031df0538043a2a2e73540d65329f96c82b98f" + } + ], + "parquet_rows": 2311, + "row_groups": 1, + "columns": [ + "image", + "split", + "prompt", + "action_id", + "action_text", + "latency", + "latency_ms", + "reward", + "done", + "episode_idx", + "t", + "seed", + "deterministic" + ], + "decoded_images": 1, + "first_row_scalars": { + "split": "train", + "action_id": 29, + "action_text": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "latency": 2, + "latency_ms": 57.142857142857146, + "reward": 0.0, + "done": false, + "episode_idx": 0, + "t": 0, + "seed": 200000, + "deterministic": true + }, + "parquet_readable": true, + "range_read_actual_data": true + }, + { + "source_repo": "latency-sensitive-bench/flappy_fixed_latency_3_200ep_7k2steps_ghost15", + "source_revision": "f90d1dac8b31a18c0496884729c744d7a4c877ae", + "source_path": "val.parquet", + "target_repo": "latency-sensitive-bench/memory-data", + "type": "dataset", + "path": "training/flappy_fixed_latency_3_200ep_7k2steps_ghost15/val.parquet", + "identity": "d8017b5d3c4480372e466f0860b69e34031b90edbc44148b95f6f28ad0aae560", + "size": 264981651, + "revision": "8a1aa27f945f1e3194764e7ef93c9993ba0db7ad", + "equivalent_paths": [ + { + "repo": "latency-sensitive-bench/memory-data", + "path": "training/flappy_fixed_latency_3_200ep_7k2steps_ghost15/val.parquet", + "revision": "bfd9e46fa650c8718200b639b7cb424c7b1014bd" + } + ], + "parquet_rows": 14400, + "row_groups": 8, + "columns": [ + "action", + "action_id", + "action_text", + "decision_step", + "env_name", + "episode_idx", + "image", + "latency_ms", + "latency_raw_frames", + "prompt", + "raw_reward", + "seed", + "split" + ], + "decoded_images": 1, + "first_row_scalars": { + "action_id": 0, + "action_text": "NOOP", + "decision_step": 0, + "env_name": "flappy", + "episode_idx": 127, + "latency_ms": 100.0, + "latency_raw_frames": 3, + "raw_reward": 0.10000000149011612, + "seed": 127, + "split": "val" + }, + "parquet_readable": true, + "range_read_actual_data": true + } + ], + "all_passed": true +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/repository_and_storage_changes.json b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/repository_and_storage_changes.json new file mode 100644 index 0000000000000000000000000000000000000000..8ef7bd6524e8ab77163b03d3edbc4fdbab2ef8da --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/reports/experiment-repos-20261007/repository_and_storage_changes.json @@ -0,0 +1,24 @@ +{ + "before": { + "model": 85, + "dataset": 98 + }, + "after": { + "model": 52, + "dataset": 53 + }, + "before_storage_bytes": 6234339659525, + "after_storage_bytes": 6149843076981, + "removed_model_repos": 34, + "removed_dataset_repos": 46, + "new_model_repos": [ + "latency-sensitive-bench/latency-information-models" + ], + "new_dataset_repos": [ + "latency-sensitive-bench/latency-information-data" + ], + "measurement": "Organization settings repository API storage, including reported repository objects/history. Server storage reclamation may finish asynchronously.", + "final_storage_bytes": 6149843076981, + "final_storage_reduction_bytes": 84496582544, + "other_repositories_storage_changes": [] +} diff --git a/experiments/latency-transfer/provenance/latency-transfer-data/run_index.json b/experiments/latency-transfer/provenance/latency-transfer-data/run_index.json new file mode 100644 index 0000000000000000000000000000000000000000..c8d472dd8f7cf522aed2509686e576b23e27b666 --- /dev/null +++ b/experiments/latency-transfer/provenance/latency-transfer-data/run_index.json @@ -0,0 +1,144 @@ +[ + { + "run": "wan_oft_flappy_fix_latency_0_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 0, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-06-28T13:44:26.129154Z", + "git_commit": "f9acb69ef3d99d2a8b620103161e15eebc4c3091", + "wandb_run_id": "ky48y0be", + "checkpoint": "wan_oft_flappy_fix_latency_0_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "daa8a61baa149139792808dcfc5406df766d39113161b92d0baa13a9bd2402f4", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_0_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 279882, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_1_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 1, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-06-29T14:50:14.198406Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "3kohuetr", + "checkpoint": "wan_oft_flappy_fix_latency_1_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "ca01daf3e577fedcc238882e531c748f6a5c178934ca7cc7209da18eb24ff03f", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_1_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 297483, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_2_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 2, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T09:28:56.645196Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "2z7am65s", + "checkpoint": "wan_oft_flappy_fix_latency_2_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "328140abd4682a66cb56cdbe03524f6c7af534e2c9ac2bd23ac1c9691363957a", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_2_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 330734, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid", + "fig13_wanoft_timing_controls", + "fig14_wanoft_timing_buckets" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_3_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 3, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T16:06:06.153675Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "jlutvxqx", + "checkpoint": "wan_oft_flappy_fix_latency_3_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "d5940d8ea363b40611f391103534da4d13dd7204e43024079c56eb2f34322bcc", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_3_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 316258, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + }, + { + "run": "wan_oft_flappy_fix_latency_4_context5_2000_effbs128", + "module": "latency-transfer", + "train_latency_raw_frames": 4, + "train_seed": 42, + "training_step": 2000, + "training_started_at_utc": "2026-07-01T17:12:40.334458Z", + "git_commit": "972037fae54ada9fe8f64e4ac8878b2f79feba1f", + "wandb_run_id": "hvokzu73", + "checkpoint": "wan_oft_flappy_fix_latency_4_context5_2000_effbs128/checkpoints/steps_2000_pytorch_model.pt", + "checkpoint_identity": "05a1c777b84de4f2b66a2db1a13eb7b1aa326167eabb30ddf50bd149d6e68118", + "source_model_revision": "1ba5d2f489f03f9cb07a637ca680fa4f2f8e51a4", + "dataset": { + "repo": "latency-sensitive-bench/latency-transfer-data", + "revision": "ecc53001af8f0019b4d50bf527c161904f27d260", + "path": "training/flappy_fix_latency_4_200ep_context5", + "source_repo": "latency-sensitive-bench/flappy_200ep_context5", + "source_revision": "19d800e5f7ec9d5311f508dbfe4a50463261616c", + "train_rows": 311588, + "train_episodes": 180, + "val_rows": 72000, + "val_episodes": 20, + "pairing_evidence": "Training local path, row count, trajectory count, action statistics and fixed raw-frame latency" + }, + "paper_consumers": [ + "fig12_wanoft_latency_grid" + ] + } +] diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_0/README.md b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_0/README.md new file mode 100644 index 0000000000000000000000000000000000000000..bf3216eb86024e0b898f3d6e7f87dd07a9fe75b3 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_0/README.md @@ -0,0 +1,46 @@ +--- +dataset_info: + features: + - name: image + dtype: image + - name: split + dtype: string + - name: prompt + dtype: string + - name: action_id + dtype: int64 + - name: action_text + dtype: string + - name: latency + dtype: int64 + - name: latency_ms + dtype: float64 + - name: reward + dtype: float64 + - name: done + dtype: bool + - name: episode_idx + dtype: int64 + - name: t + dtype: int64 + - name: seed + dtype: int64 + - name: deterministic + dtype: bool + splits: + - name: train + num_bytes: 65115523 + num_examples: 3836 + - name: val + num_bytes: 7227473 + num_examples: 443 + download_size: 67742624 + dataset_size: 72342996 +configs: +- config_name: default + data_files: + - split: train + path: data/train-* + - split: val + path: data/val-* +--- diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_0/metadata.json b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_0/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..2022590c269191ca18931126ed2bcad3358caa01 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_0/metadata.json @@ -0,0 +1,83 @@ +{ + "action_labels": { + "0": { + "0": "NOOP", + "1": "ATTACK", + "2": "MOVE_LEFT", + "3": "MOVE_LEFT + ATTACK", + "4": "MOVE_RIGHT", + "5": "MOVE_RIGHT + ATTACK", + "6": "MOVE_FORWARD", + "7": "MOVE_FORWARD + ATTACK", + "8": "MOVE_FORWARD + MOVE_LEFT", + "9": "MOVE_FORWARD + MOVE_LEFT + ATTACK", + "10": "MOVE_FORWARD + MOVE_RIGHT", + "11": "MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "12": "MOVE_BACKWARD", + "13": "MOVE_BACKWARD + ATTACK", + "14": "MOVE_BACKWARD + MOVE_LEFT", + "15": "MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "16": "MOVE_BACKWARD + MOVE_RIGHT", + "17": "MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "18": "TURN_LEFT", + "19": "TURN_LEFT + ATTACK", + "20": "TURN_LEFT + MOVE_LEFT", + "21": "TURN_LEFT + MOVE_LEFT + ATTACK", + "22": "TURN_LEFT + MOVE_RIGHT", + "23": "TURN_LEFT + MOVE_RIGHT + ATTACK", + "24": "TURN_LEFT + MOVE_FORWARD", + "25": "TURN_LEFT + MOVE_FORWARD + ATTACK", + "26": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "27": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "28": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "29": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "30": "TURN_LEFT + MOVE_BACKWARD", + "31": "TURN_LEFT + MOVE_BACKWARD + ATTACK", + "32": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT", + "33": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "34": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "35": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "36": "TURN_RIGHT", + "37": "TURN_RIGHT + ATTACK", + "38": "TURN_RIGHT + MOVE_LEFT", + "39": "TURN_RIGHT + MOVE_LEFT + ATTACK", + "40": "TURN_RIGHT + MOVE_RIGHT", + "41": "TURN_RIGHT + MOVE_RIGHT + ATTACK", + "42": "TURN_RIGHT + MOVE_FORWARD", + "43": "TURN_RIGHT + MOVE_FORWARD + ATTACK", + "44": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT", + "45": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "46": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT", + "47": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "48": "TURN_RIGHT + MOVE_BACKWARD", + "49": "TURN_RIGHT + MOVE_BACKWARD + ATTACK", + "50": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT", + "51": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "52": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT", + "53": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK" + } + }, + "checkpoint_root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints", + "dataset_mode": "fixed_latency", + "deterministic": true, + "env_name": "deadly_corridor", + "export_image_size": null, + "image_source": "render", + "image_writer_workers": 32, + "latencies": [ + { + "best_checkpoint": "best_000012896_26411008_reward_2220.457.pth", + "episode_count": 100, + "experiment_dir": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints/deadly_corridor_sf_same_latency_l0_seed0", + "latency": 0, + "latency_ms": 0.0, + "val_episode_count": 10 + } + ], + "latency_unit": "raw_frames", + "max_steps_per_episode": 2000, + "mixed_prompt_template": "{prompt} The current action latency is {latency} raw frames ({latency_ms} ms). Choose the best next action while accounting for the delay.", + "num_envs": 64, + "prompt": "You are playing Deadly Corridor in VizDoom. Your goal is to survive the corridor, fight enemies, and reach the green armor vest at the far end. The available actions are: MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, and ATTACK. Choose the best next action from the current visual observation.", + "row_buffer_size": 8192 +} diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/README.md b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/README.md new file mode 100644 index 0000000000000000000000000000000000000000..069b579669d0b972c9328290e3d3ee347e392a21 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/README.md @@ -0,0 +1,46 @@ +--- +dataset_info: + features: + - name: image + dtype: image + - name: split + dtype: string + - name: prompt + dtype: string + - name: action_id + dtype: int64 + - name: action_text + dtype: string + - name: latency + dtype: int64 + - name: latency_ms + dtype: float64 + - name: reward + dtype: float64 + - name: done + dtype: bool + - name: episode_idx + dtype: int64 + - name: t + dtype: int64 + - name: seed + dtype: int64 + - name: deterministic + dtype: bool + splits: + - name: train + num_bytes: 34549455 + num_examples: 1966 + - name: val + num_bytes: 3750078 + num_examples: 218 + download_size: 36651540 + dataset_size: 38299533 +configs: +- config_name: default + data_files: + - split: train + path: data/train-* + - split: val + path: data/val-* +--- diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/data/train-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/data/train-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..ceb53e858c5fe009e9f549744cbedb7ad0a392bb --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/data/train-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:893c4234419bcce925c4e739d97c20b411fb1bf2cbbf30d085009195cefe86a2 +size 32981127 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/data/val-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/data/val-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..abb07d2091815f8c22b5d4e82e6a17401c0646bd --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/data/val-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab8353529f419e970833e4e87ca3c0ccad5dd3e679e461e3472eccadc943b729 +size 3670413 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/metadata.json b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..1a391ee63a9e65cd3931d7326e98bacb7b8290c4 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_1/metadata.json @@ -0,0 +1,83 @@ +{ + "action_labels": { + "1": { + "0": "NOOP", + "1": "ATTACK", + "2": "MOVE_LEFT", + "3": "MOVE_LEFT + ATTACK", + "4": "MOVE_RIGHT", + "5": "MOVE_RIGHT + ATTACK", + "6": "MOVE_FORWARD", + "7": "MOVE_FORWARD + ATTACK", + "8": "MOVE_FORWARD + MOVE_LEFT", + "9": "MOVE_FORWARD + MOVE_LEFT + ATTACK", + "10": "MOVE_FORWARD + MOVE_RIGHT", + "11": "MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "12": "MOVE_BACKWARD", + "13": "MOVE_BACKWARD + ATTACK", + "14": "MOVE_BACKWARD + MOVE_LEFT", + "15": "MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "16": "MOVE_BACKWARD + MOVE_RIGHT", + "17": "MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "18": "TURN_LEFT", + "19": "TURN_LEFT + ATTACK", + "20": "TURN_LEFT + MOVE_LEFT", + "21": "TURN_LEFT + MOVE_LEFT + ATTACK", + "22": "TURN_LEFT + MOVE_RIGHT", + "23": "TURN_LEFT + MOVE_RIGHT + ATTACK", + "24": "TURN_LEFT + MOVE_FORWARD", + "25": "TURN_LEFT + MOVE_FORWARD + ATTACK", + "26": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "27": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "28": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "29": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "30": "TURN_LEFT + MOVE_BACKWARD", + "31": "TURN_LEFT + MOVE_BACKWARD + ATTACK", + "32": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT", + "33": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "34": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "35": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "36": "TURN_RIGHT", + "37": "TURN_RIGHT + ATTACK", + "38": "TURN_RIGHT + MOVE_LEFT", + "39": "TURN_RIGHT + MOVE_LEFT + ATTACK", + "40": "TURN_RIGHT + MOVE_RIGHT", + "41": "TURN_RIGHT + MOVE_RIGHT + ATTACK", + "42": "TURN_RIGHT + MOVE_FORWARD", + "43": "TURN_RIGHT + MOVE_FORWARD + ATTACK", + "44": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT", + "45": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "46": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT", + "47": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "48": "TURN_RIGHT + MOVE_BACKWARD", + "49": "TURN_RIGHT + MOVE_BACKWARD + ATTACK", + "50": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT", + "51": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "52": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT", + "53": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK" + } + }, + "checkpoint_root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints", + "dataset_mode": "fixed_latency", + "deterministic": true, + "env_name": "deadly_corridor", + "export_image_size": null, + "image_source": "render", + "image_writer_workers": 32, + "latencies": [ + { + "best_checkpoint": "best_000023416_47955968_reward_1989.190.pth", + "episode_count": 100, + "experiment_dir": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints/deadly_corridor_sf_same_latency_l1_seed0", + "latency": 1, + "latency_ms": 28.571428571428573, + "val_episode_count": 10 + } + ], + "latency_unit": "raw_frames", + "max_steps_per_episode": 2000, + "mixed_prompt_template": "{prompt} The current action latency is {latency} raw frames ({latency_ms} ms). Choose the best next action while accounting for the delay.", + "num_envs": 64, + "prompt": "You are playing Deadly Corridor in VizDoom. Your goal is to survive the corridor, fight enemies, and reach the green armor vest at the far end. The available actions are: MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, and ATTACK. Choose the best next action from the current visual observation.", + "row_buffer_size": 8192 +} diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/README.md b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/README.md new file mode 100644 index 0000000000000000000000000000000000000000..0f51aa1d25345a5d563e25367a389f7a7592b2e8 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/README.md @@ -0,0 +1,46 @@ +--- +dataset_info: + features: + - name: image + dtype: image + - name: split + dtype: string + - name: prompt + dtype: string + - name: action_id + dtype: int64 + - name: action_text + dtype: string + - name: latency + dtype: int64 + - name: latency_ms + dtype: float64 + - name: reward + dtype: float64 + - name: done + dtype: bool + - name: episode_idx + dtype: int64 + - name: t + dtype: int64 + - name: seed + dtype: int64 + - name: deterministic + dtype: bool + splits: + - name: train + num_bytes: 33223147 + num_examples: 1989 + - name: val + num_bytes: 4470488 + num_examples: 271 + download_size: 36553730 + dataset_size: 37693635 +configs: +- config_name: default + data_files: + - split: train + path: data/train-* + - split: val + path: data/val-* +--- diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/data/train-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/data/train-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..1895fa085ac6f3fb06f0d4aedfd94ee3da0ff236 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/data/train-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be218fed0bb91c69feee31e7eb1ca2a3b6e0c15b3d96f918da26c5b73e455442 +size 32185015 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/data/val-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/data/val-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..0a181ca42d6476083ab6802ed61e6c7972d346ba --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/data/val-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:25854a5f15dfcdcd3405bc549f238bbe0b7af55aa15f041b82870c911a19be5c +size 4368715 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/metadata.json b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..4f6a0550466015ec1b326acbaf018fba0238e286 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_10/metadata.json @@ -0,0 +1,86 @@ +{ + "action_labels": { + "10": { + "0": "NOOP", + "1": "ATTACK", + "2": "MOVE_LEFT", + "3": "MOVE_LEFT + ATTACK", + "4": "MOVE_RIGHT", + "5": "MOVE_RIGHT + ATTACK", + "6": "MOVE_FORWARD", + "7": "MOVE_FORWARD + ATTACK", + "8": "MOVE_FORWARD + MOVE_LEFT", + "9": "MOVE_FORWARD + MOVE_LEFT + ATTACK", + "10": "MOVE_FORWARD + MOVE_RIGHT", + "11": "MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "12": "MOVE_BACKWARD", + "13": "MOVE_BACKWARD + ATTACK", + "14": "MOVE_BACKWARD + MOVE_LEFT", + "15": "MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "16": "MOVE_BACKWARD + MOVE_RIGHT", + "17": "MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "18": "TURN_LEFT", + "19": "TURN_LEFT + ATTACK", + "20": "TURN_LEFT + MOVE_LEFT", + "21": "TURN_LEFT + MOVE_LEFT + ATTACK", + "22": "TURN_LEFT + MOVE_RIGHT", + "23": "TURN_LEFT + MOVE_RIGHT + ATTACK", + "24": "TURN_LEFT + MOVE_FORWARD", + "25": "TURN_LEFT + MOVE_FORWARD + ATTACK", + "26": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "27": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "28": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "29": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "30": "TURN_LEFT + MOVE_BACKWARD", + "31": "TURN_LEFT + MOVE_BACKWARD + ATTACK", + "32": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT", + "33": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "34": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "35": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "36": "TURN_RIGHT", + "37": "TURN_RIGHT + ATTACK", + "38": "TURN_RIGHT + MOVE_LEFT", + "39": "TURN_RIGHT + MOVE_LEFT + ATTACK", + "40": "TURN_RIGHT + MOVE_RIGHT", + "41": "TURN_RIGHT + MOVE_RIGHT + ATTACK", + "42": "TURN_RIGHT + MOVE_FORWARD", + "43": "TURN_RIGHT + MOVE_FORWARD + ATTACK", + "44": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT", + "45": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "46": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT", + "47": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "48": "TURN_RIGHT + MOVE_BACKWARD", + "49": "TURN_RIGHT + MOVE_BACKWARD + ATTACK", + "50": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT", + "51": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "52": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT", + "53": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK" + } + }, + "checkpoint_root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints", + "dataset_mode": "fixed_latency", + "deterministic": true, + "env_name": "deadly_corridor", + "export_image_size": null, + "image_source": "render", + "image_writer_workers": 32, + "latencies": [ + { + "best_checkpoint": "best_000024256_49676288_reward_1258.361.pth", + "collected_episode_count": 100, + "collected_train_episode_count": 90, + "collected_val_episode_count": 10, + "episode_count": 100, + "experiment_dir": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints/deadly_corridor_sf_same_latency_l10_seed0", + "latency": 10, + "latency_ms": 285.7142857142857, + "val_episode_count": 10 + } + ], + "latency_unit": "raw_frames", + "max_steps_per_episode": 2000, + "mixed_prompt_template": "{prompt} The current action latency is {latency} raw frames ({latency_ms} ms). Choose the best next action while accounting for the delay.", + "num_envs": 64, + "prompt": "You are playing Deadly Corridor in VizDoom. Your goal is to survive the corridor, fight enemies, and reach the green armor vest at the far end. The available actions are: MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, and ATTACK. Choose the best next action from the current visual observation.", + "row_buffer_size": 8192 +} diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/README.md b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/README.md new file mode 100644 index 0000000000000000000000000000000000000000..517c494e0cc1a427c186dd7da2d58d87f62ae851 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/README.md @@ -0,0 +1,46 @@ +--- +dataset_info: + features: + - name: image + dtype: image + - name: split + dtype: string + - name: prompt + dtype: string + - name: action_id + dtype: int64 + - name: action_text + dtype: string + - name: latency + dtype: int64 + - name: latency_ms + dtype: float64 + - name: reward + dtype: float64 + - name: done + dtype: bool + - name: episode_idx + dtype: int64 + - name: t + dtype: int64 + - name: seed + dtype: int64 + - name: deterministic + dtype: bool + splits: + - name: train + num_bytes: 42932075 + num_examples: 2501 + - name: val + num_bytes: 4242789 + num_examples: 253 + download_size: 44786426 + dataset_size: 47174864 +configs: +- config_name: default + data_files: + - split: train + path: data/train-* + - split: val + path: data/val-* +--- diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/data/train-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/data/train-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..e59d34ed575cb8c95746912632fab5c889a0a3cf --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/data/train-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c512c87ab13f1fcde4dd1cd013361d0e2f886b1c895df5bb9e163bc47a821505 +size 40638421 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/data/val-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/data/val-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..2afa7c542a357ee79d2d52b3c09fff7a74a25aa9 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/data/val-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7e49bd90866da520bf9120a16ec9aa197cd493e5c3ba79379dbbb44dfa72c8e +size 4148005 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/metadata.json b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..1170a934b14628704470ffacd5b5b7e4f1ef259f --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_12/metadata.json @@ -0,0 +1,86 @@ +{ + "action_labels": { + "12": { + "0": "NOOP", + "1": "ATTACK", + "2": "MOVE_LEFT", + "3": "MOVE_LEFT + ATTACK", + "4": "MOVE_RIGHT", + "5": "MOVE_RIGHT + ATTACK", + "6": "MOVE_FORWARD", + "7": "MOVE_FORWARD + ATTACK", + "8": "MOVE_FORWARD + MOVE_LEFT", + "9": "MOVE_FORWARD + MOVE_LEFT + ATTACK", + "10": "MOVE_FORWARD + MOVE_RIGHT", + "11": "MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "12": "MOVE_BACKWARD", + "13": "MOVE_BACKWARD + ATTACK", + "14": "MOVE_BACKWARD + MOVE_LEFT", + "15": "MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "16": "MOVE_BACKWARD + MOVE_RIGHT", + "17": "MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "18": "TURN_LEFT", + "19": "TURN_LEFT + ATTACK", + "20": "TURN_LEFT + MOVE_LEFT", + "21": "TURN_LEFT + MOVE_LEFT + ATTACK", + "22": "TURN_LEFT + MOVE_RIGHT", + "23": "TURN_LEFT + MOVE_RIGHT + ATTACK", + "24": "TURN_LEFT + MOVE_FORWARD", + "25": "TURN_LEFT + MOVE_FORWARD + ATTACK", + "26": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "27": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "28": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "29": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "30": "TURN_LEFT + MOVE_BACKWARD", + "31": "TURN_LEFT + MOVE_BACKWARD + ATTACK", + "32": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT", + "33": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "34": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "35": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "36": "TURN_RIGHT", + "37": "TURN_RIGHT + ATTACK", + "38": "TURN_RIGHT + MOVE_LEFT", + "39": "TURN_RIGHT + MOVE_LEFT + ATTACK", + "40": "TURN_RIGHT + MOVE_RIGHT", + "41": "TURN_RIGHT + MOVE_RIGHT + ATTACK", + "42": "TURN_RIGHT + MOVE_FORWARD", + "43": "TURN_RIGHT + MOVE_FORWARD + ATTACK", + "44": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT", + "45": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "46": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT", + "47": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "48": "TURN_RIGHT + MOVE_BACKWARD", + "49": "TURN_RIGHT + MOVE_BACKWARD + ATTACK", + "50": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT", + "51": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "52": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT", + "53": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK" + } + }, + "checkpoint_root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints", + "dataset_mode": "fixed_latency", + "deterministic": true, + "env_name": "deadly_corridor", + "export_image_size": null, + "image_source": "render", + "image_writer_workers": 32, + "latencies": [ + { + "best_checkpoint": "best_000023868_48881664_reward_1028.417.pth", + "collected_episode_count": 100, + "collected_train_episode_count": 90, + "collected_val_episode_count": 10, + "episode_count": 100, + "experiment_dir": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints/deadly_corridor_sf_same_latency_l12_seed0", + "latency": 12, + "latency_ms": 342.85714285714283, + "val_episode_count": 10 + } + ], + "latency_unit": "raw_frames", + "max_steps_per_episode": 2000, + "mixed_prompt_template": "{prompt} The current action latency is {latency} raw frames ({latency_ms} ms). Choose the best next action while accounting for the delay.", + "num_envs": 64, + "prompt": "You are playing Deadly Corridor in VizDoom. Your goal is to survive the corridor, fight enemies, and reach the green armor vest at the far end. The available actions are: MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, and ATTACK. Choose the best next action from the current visual observation.", + "row_buffer_size": 8192 +} diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/README.md b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/README.md new file mode 100644 index 0000000000000000000000000000000000000000..10d4e342799813071b22fd3d15e8b76f592aa54c --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/README.md @@ -0,0 +1,46 @@ +--- +dataset_info: + features: + - name: image + dtype: image + - name: split + dtype: string + - name: prompt + dtype: string + - name: action_id + dtype: int64 + - name: action_text + dtype: string + - name: latency + dtype: int64 + - name: latency_ms + dtype: float64 + - name: reward + dtype: float64 + - name: done + dtype: bool + - name: episode_idx + dtype: int64 + - name: t + dtype: int64 + - name: seed + dtype: int64 + - name: deterministic + dtype: bool + splits: + - name: train + num_bytes: 41906524 + num_examples: 2442 + - name: val + num_bytes: 4480805 + num_examples: 268 + download_size: 43609149 + dataset_size: 46387329 +configs: +- config_name: default + data_files: + - split: train + path: data/train-* + - split: val + path: data/val-* +--- diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/data/train-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/data/train-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..38d3462876b4f36d0a4ef70f9426397e617d7873 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/data/train-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3a51e43947c9d2c66437fa8a4acbcf3c05abc3dc4c1f87b7f63ab8c703bed96 +size 39227814 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/data/val-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/data/val-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..ebed6aa026c89a9bdd79a191e1b2e888f7cad344 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/data/val-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf22ceaf3817f67308a7ade800567aed5b341c06b9c1255f63afd2534b937f35 +size 4381335 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/metadata.json b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..c7af2e06dbad7bbcfcc53f1fb6114660cfef5414 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_13/metadata.json @@ -0,0 +1,86 @@ +{ + "action_labels": { + "13": { + "0": "NOOP", + "1": "ATTACK", + "2": "MOVE_LEFT", + "3": "MOVE_LEFT + ATTACK", + "4": "MOVE_RIGHT", + "5": "MOVE_RIGHT + ATTACK", + "6": "MOVE_FORWARD", + "7": "MOVE_FORWARD + ATTACK", + "8": "MOVE_FORWARD + MOVE_LEFT", + "9": "MOVE_FORWARD + MOVE_LEFT + ATTACK", + "10": "MOVE_FORWARD + MOVE_RIGHT", + "11": "MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "12": "MOVE_BACKWARD", + "13": "MOVE_BACKWARD + ATTACK", + "14": "MOVE_BACKWARD + MOVE_LEFT", + "15": "MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "16": "MOVE_BACKWARD + MOVE_RIGHT", + "17": "MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "18": "TURN_LEFT", + "19": "TURN_LEFT + ATTACK", + "20": "TURN_LEFT + MOVE_LEFT", + "21": "TURN_LEFT + MOVE_LEFT + ATTACK", + "22": "TURN_LEFT + MOVE_RIGHT", + "23": "TURN_LEFT + MOVE_RIGHT + ATTACK", + "24": "TURN_LEFT + MOVE_FORWARD", + "25": "TURN_LEFT + MOVE_FORWARD + ATTACK", + "26": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "27": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "28": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "29": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "30": "TURN_LEFT + MOVE_BACKWARD", + "31": "TURN_LEFT + MOVE_BACKWARD + ATTACK", + "32": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT", + "33": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "34": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "35": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "36": "TURN_RIGHT", + "37": "TURN_RIGHT + ATTACK", + "38": "TURN_RIGHT + MOVE_LEFT", + "39": "TURN_RIGHT + MOVE_LEFT + ATTACK", + "40": "TURN_RIGHT + MOVE_RIGHT", + "41": "TURN_RIGHT + MOVE_RIGHT + ATTACK", + "42": "TURN_RIGHT + MOVE_FORWARD", + "43": "TURN_RIGHT + MOVE_FORWARD + ATTACK", + "44": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT", + "45": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "46": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT", + "47": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "48": "TURN_RIGHT + MOVE_BACKWARD", + "49": "TURN_RIGHT + MOVE_BACKWARD + ATTACK", + "50": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT", + "51": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "52": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT", + "53": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK" + } + }, + "checkpoint_root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints", + "dataset_mode": "fixed_latency", + "deterministic": true, + "env_name": "deadly_corridor", + "export_image_size": null, + "image_source": "render", + "image_writer_workers": 32, + "latencies": [ + { + "best_checkpoint": "best_000024136_49430528_reward_575.345.pth", + "collected_episode_count": 100, + "collected_train_episode_count": 90, + "collected_val_episode_count": 10, + "episode_count": 100, + "experiment_dir": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints/deadly_corridor_sf_same_latency_l13_seed0", + "latency": 13, + "latency_ms": 371.42857142857144, + "val_episode_count": 10 + } + ], + "latency_unit": "raw_frames", + "max_steps_per_episode": 2000, + "mixed_prompt_template": "{prompt} The current action latency is {latency} raw frames ({latency_ms} ms). Choose the best next action while accounting for the delay.", + "num_envs": 64, + "prompt": "You are playing Deadly Corridor in VizDoom. Your goal is to survive the corridor, fight enemies, and reach the green armor vest at the far end. The available actions are: MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, and ATTACK. Choose the best next action from the current visual observation.", + "row_buffer_size": 8192 +} diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/README.md b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/README.md new file mode 100644 index 0000000000000000000000000000000000000000..30779ff28e32f05f00ff0fedf74a4de71f5ee05b --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/README.md @@ -0,0 +1,46 @@ +--- +dataset_info: + features: + - name: image + dtype: image + - name: split + dtype: string + - name: prompt + dtype: string + - name: action_id + dtype: int64 + - name: action_text + dtype: string + - name: latency + dtype: int64 + - name: latency_ms + dtype: float64 + - name: reward + dtype: float64 + - name: done + dtype: bool + - name: episode_idx + dtype: int64 + - name: t + dtype: int64 + - name: seed + dtype: int64 + - name: deterministic + dtype: bool + splits: + - name: train + num_bytes: 38448876 + num_examples: 2325 + - name: val + num_bytes: 3844979 + num_examples: 238 + download_size: 40536512 + dataset_size: 42293855 +configs: +- config_name: default + data_files: + - split: train + path: data/train-* + - split: val + path: data/val-* +--- diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/data/train-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/data/train-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..1979b303838a9b4f28ae11cfe0848a65c21bf66b --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/data/train-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d8bc75ef45e8a4f6755450c43cf408d3c0d857e5fed06abf9642cac43ec1c90 +size 36779288 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/data/val-00000-of-00001.parquet b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/data/val-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..a685b75fb38ffd91f60d65531ba2ad7855ff4506 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/data/val-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cb0a856d06b6d1e0ff53b995b7dfccc370111edaa2d221bcdf05a2150ec3de6 +size 3757224 diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/metadata.json b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/metadata.json new file mode 100644 index 0000000000000000000000000000000000000000..4f2720041f6e8d7db0ca7223ca0e3ab0bca622c4 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_14/metadata.json @@ -0,0 +1,86 @@ +{ + "action_labels": { + "14": { + "0": "NOOP", + "1": "ATTACK", + "2": "MOVE_LEFT", + "3": "MOVE_LEFT + ATTACK", + "4": "MOVE_RIGHT", + "5": "MOVE_RIGHT + ATTACK", + "6": "MOVE_FORWARD", + "7": "MOVE_FORWARD + ATTACK", + "8": "MOVE_FORWARD + MOVE_LEFT", + "9": "MOVE_FORWARD + MOVE_LEFT + ATTACK", + "10": "MOVE_FORWARD + MOVE_RIGHT", + "11": "MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "12": "MOVE_BACKWARD", + "13": "MOVE_BACKWARD + ATTACK", + "14": "MOVE_BACKWARD + MOVE_LEFT", + "15": "MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "16": "MOVE_BACKWARD + MOVE_RIGHT", + "17": "MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "18": "TURN_LEFT", + "19": "TURN_LEFT + ATTACK", + "20": "TURN_LEFT + MOVE_LEFT", + "21": "TURN_LEFT + MOVE_LEFT + ATTACK", + "22": "TURN_LEFT + MOVE_RIGHT", + "23": "TURN_LEFT + MOVE_RIGHT + ATTACK", + "24": "TURN_LEFT + MOVE_FORWARD", + "25": "TURN_LEFT + MOVE_FORWARD + ATTACK", + "26": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT", + "27": "TURN_LEFT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "28": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT", + "29": "TURN_LEFT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "30": "TURN_LEFT + MOVE_BACKWARD", + "31": "TURN_LEFT + MOVE_BACKWARD + ATTACK", + "32": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT", + "33": "TURN_LEFT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "34": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT", + "35": "TURN_LEFT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK", + "36": "TURN_RIGHT", + "37": "TURN_RIGHT + ATTACK", + "38": "TURN_RIGHT + MOVE_LEFT", + "39": "TURN_RIGHT + MOVE_LEFT + ATTACK", + "40": "TURN_RIGHT + MOVE_RIGHT", + "41": "TURN_RIGHT + MOVE_RIGHT + ATTACK", + "42": "TURN_RIGHT + MOVE_FORWARD", + "43": "TURN_RIGHT + MOVE_FORWARD + ATTACK", + "44": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT", + "45": "TURN_RIGHT + MOVE_FORWARD + MOVE_LEFT + ATTACK", + "46": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT", + "47": "TURN_RIGHT + MOVE_FORWARD + MOVE_RIGHT + ATTACK", + "48": "TURN_RIGHT + MOVE_BACKWARD", + "49": "TURN_RIGHT + MOVE_BACKWARD + ATTACK", + "50": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT", + "51": "TURN_RIGHT + MOVE_BACKWARD + MOVE_LEFT + ATTACK", + "52": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT", + "53": "TURN_RIGHT + MOVE_BACKWARD + MOVE_RIGHT + ATTACK" + } + }, + "checkpoint_root": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints", + "dataset_mode": "fixed_latency", + "deterministic": true, + "env_name": "deadly_corridor", + "export_image_size": null, + "image_source": "render", + "image_writer_workers": 32, + "latencies": [ + { + "best_checkpoint": "best_000024188_49537024_reward_356.299.pth", + "collected_episode_count": 100, + "collected_train_episode_count": 90, + "collected_val_episode_count": 10, + "episode_count": 100, + "experiment_dir": "/inspire/hdd/global_user/liumingyu-253208120284/lzj/latency-sensitive-bench/checkpoints/deadly_corridor_sf_same_latency_l14_seed0", + "latency": 14, + "latency_ms": 400.0, + "val_episode_count": 10 + } + ], + "latency_unit": "raw_frames", + "max_steps_per_episode": 2000, + "mixed_prompt_template": "{prompt} The current action latency is {latency} raw frames ({latency_ms} ms). Choose the best next action while accounting for the delay.", + "num_envs": 64, + "prompt": "You are playing Deadly Corridor in VizDoom. Your goal is to survive the corridor, fight enemies, and reach the green armor vest at the far end. The available actions are: MOVE_FORWARD, MOVE_BACKWARD, MOVE_LEFT, MOVE_RIGHT, TURN_LEFT, TURN_RIGHT, and ATTACK. Choose the best next action from the current visual observation.", + "row_buffer_size": 8192 +} diff --git a/latency-aware/deadly-corridor/deadly_corridor_fix_latency_15/README.md b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_15/README.md new file mode 100644 index 0000000000000000000000000000000000000000..dc71cabfea90075987def4813954404078a0c571 --- /dev/null +++ b/latency-aware/deadly-corridor/deadly_corridor_fix_latency_15/README.md @@ -0,0 +1,46 @@ +--- +dataset_info: + features: + - name: image + dtype: image + - name: split + dtype: string + - name: prompt + dtype: string + - name: action_id + dtype: int64 + - name: action_text + dtype: string + - name: latency + dtype: int64 + - name: latency_ms + dtype: float64 + - name: reward + dtype: float64 + - name: done + dtype: bool + - name: episode_idx + dtype: int64 + - name: t + dtype: int64 + - name: seed + dtype: int64 + - name: deterministic + dtype: bool + splits: + - name: train + num_bytes: 33416481 + num_examples: 2061 + - name: val + num_bytes: 4101010 + num_examples: 259 + download_size: 36304256 + dataset_size: 37517491 +configs: +- config_name: default + data_files: + - split: train + path: data/train-* + - split: val + path: data/val-* +--- diff --git a/zero-latency/deadly-corridor/deadly_corridor_fix_latency_0/data/train-00000-of-00001.parquet b/zero-latency/deadly-corridor/deadly_corridor_fix_latency_0/data/train-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..961181a2c89a042abc215d0415b3520ba6ce617b --- /dev/null +++ b/zero-latency/deadly-corridor/deadly_corridor_fix_latency_0/data/train-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7e5c9a1b249cce02b696b25d7069c094ff14d54b73da87b49a5b74f98115c03 +size 60683500 diff --git a/zero-latency/deadly-corridor/deadly_corridor_fix_latency_0/data/val-00000-of-00001.parquet b/zero-latency/deadly-corridor/deadly_corridor_fix_latency_0/data/val-00000-of-00001.parquet new file mode 100644 index 0000000000000000000000000000000000000000..59ead584f0a93dd90346a222694e1453040318f5 --- /dev/null +++ b/zero-latency/deadly-corridor/deadly_corridor_fix_latency_0/data/val-00000-of-00001.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0931d75b4b291662663d0703150915e6f7e60408c7e836e7dc3358df8a7d3a6b +size 7059124