{
  "completed_at": "2025-01-06T20:39:56.629216Z",
  "created_at": "2025-01-06T19:04:53.153000Z",
  "data_removed": false,
  "error": null,
  "id": "6cdr553mm5rm80cm7m0vk6bhsr",
  "input": {
    "rank": 128,
    "seed": 0,
    "hf_token": "[REDACTED]",
    "batch_size": 1,
    "train_steps": 200,
    "hub_model_id": "lucataco/hunyuan-lora-melty-test-3",
    "input_videos": "https://replicate.delivery/pbxt/MHANihgtbQjWzg3dY8V0YsHcVZMqaSo3bVy75HxMDeiJjIgN/melty-3.zip",
    "trigger_word": "melty",
    "gradient_accumulation_steps": 1
  },
  "logs": "Using seed: 58933\nCleaning up past runs\nExtracted 5 files from zip to folder: dataset\nUsing args: ['accelerate', 'launch', '--config_file', 'accelerate_configs/uncompiled_1.yaml', '--gpu_ids', '0', 'train.py', '--model_name', 'hunyuan_video', '--pretrained_model_name_or_path', 'HunyuanVideo', '--enable_tiling', '--enable_slicing', '--data_root', 'dataset', '--caption_column', 'prompts.txt', '--video_column', 'videos.txt', '--seed', '58933', '--rank', '128', '--lora_alpha', '128', '--mixed_precision', 'bf16', '--output_dir', 'output', '--batch_size', '1', '--id_token', 'melty', '--video_resolution_buckets', '17x512x768', '49x512x768', '61x512x768', '--caption_dropout_p', '0.05', '--training_type', 'lora', '--train_steps', '200', '--gradient_accumulation_steps', '1', '--gradient_checkpointing', '--optimizer', 'adamw', '--lr', '2e-05', '--lr_scheduler', 'constant_with_warmup', '--lr_warmup_steps', '100', '--lr_num_cycles', '1', '--beta1', '0.9', '--beta2', '0.95', '--max_grad_norm', '1.0', '--weight_decay', '0.0001', '--epsilon', '1e-08', '--tracker_name', 'replicate-hunyuanvideo']\n01/06/2025 19:06:10 - INFO - finetrainers - Initialized FineTrainers\n01/06/2025 19:06:10 - INFO - finetrainers - Distributed environment: DistributedType.NO\nNum processes: 1\nProcess index: 0\nLocal process index: 0\nDevice: cuda\nMixed precision type: bf16\n01/06/2025 19:06:10 - INFO - finetrainers - Initializing dataset and dataloader\n01/06/2025 19:06:10 - INFO - finetrainers - Initializing models\nLoading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]\nLoading checkpoint shards:  25%|██▌       | 1/4 [00:04<00:13,  4.38s/it]\nLoading checkpoint shards:  50%|█████     | 2/4 [00:08<00:08,  4.33s/it]\nLoading checkpoint shards:  75%|███████▌  | 3/4 [00:12<00:04,  4.22s/it]\nLoading checkpoint shards: 100%|██████████| 4/4 [00:12<00:00,  2.64s/it]\nLoading checkpoint shards: 100%|██████████| 4/4 [00:12<00:00,  3.25s/it]\n01/06/2025 19:06:25 - INFO - finetrainers - Initializing trainable parameters\n01/06/2025 19:06:33 - INFO - finetrainers - Initializing optimizer and lr scheduler\n01/06/2025 19:06:33 - INFO - finetrainers - Initializing trackers\n01/06/2025 19:06:33 - INFO - finetrainers - Starting training\n01/06/2025 19:06:33 - INFO - finetrainers - Memory before training start: {\n\"memory_allocated\": 38.889,\n\"memory_reserved\": 39.02,\n\"max_memory_allocated\": 38.889,\n\"max_memory_reserved\": 39.02\n}\n01/06/2025 19:06:33 - INFO - finetrainers - Training configuration: {\n\"trainable parameters\": 163577856,\n\"total samples\": 3,\n\"train epochs\": 67,\n\"train steps\": 200,\n\"batches per device\": 1,\n\"total batches observed per epoch\": 3,\n\"train batch size\": 1,\n\"gradient accumulation steps\": 1\n}\nTraining steps:   0%|          | 0/200 [00:00<?, ?it/s]\nTraining steps:   0%|          | 1/200 [00:32<1:47:26, 32.39s/it]\nTraining steps:   0%|          | 1/200 [00:32<1:47:26, 32.39s/it, grad_norm=0.0184, loss=0.165, lr=2e-7]\nTraining steps:   1%|          | 2/200 [01:00<1:37:37, 29.58s/it, grad_norm=0.0184, loss=0.165, lr=2e-7]\nTraining steps:   1%|          | 2/200 [01:00<1:37:37, 29.58s/it, grad_norm=0.00458, loss=0.381, lr=4e-7]\nTraining steps:   2%|▏         | 3/200 [01:25<1:30:54, 27.69s/it, grad_norm=0.00458, loss=0.381, lr=4e-7]\nTraining steps:   2%|▏         | 3/200 [01:25<1:30:54, 27.69s/it, grad_norm=0.00131, loss=0.0621, lr=6e-7]\nTraining steps:   2%|▏         | 4/200 [01:57<1:35:34, 29.26s/it, grad_norm=0.00131, loss=0.0621, lr=6e-7]\nTraining steps:   2%|▏         | 4/200 [01:57<1:35:34, 29.26s/it, grad_norm=0.00873, loss=0.561, lr=8e-7] \nTraining steps:   2%|▎         | 5/200 [02:23<1:31:49, 28.26s/it, grad_norm=0.00873, loss=0.561, lr=8e-7]\nTraining steps:   2%|▎         | 5/200 [02:23<1:31:49, 28.26s/it, grad_norm=0.00436, loss=0.392, lr=1e-6]\nTraining steps:   3%|▎         | 6/200 [02:49<1:28:15, 27.30s/it, grad_norm=0.00436, loss=0.392, lr=1e-6]\nTraining steps:   3%|▎         | 6/200 [02:49<1:28:15, 27.30s/it, grad_norm=0.0104, loss=0.541, lr=1.2e-6]\nTraining steps:   4%|▎         | 7/200 [03:20<1:32:20, 28.71s/it, grad_norm=0.0104, loss=0.541, lr=1.2e-6]\nTraining steps:   4%|▎         | 7/200 [03:20<1:32:20, 28.71s/it, grad_norm=0.00531, loss=0.299, lr=1.4e-6]\nTraining steps:   4%|▍         | 8/200 [03:47<1:29:33, 27.99s/it, grad_norm=0.00531, loss=0.299, lr=1.4e-6]\nTraining steps:   4%|▍         | 8/200 [03:47<1:29:33, 27.99s/it, grad_norm=0.00241, loss=0.114, lr=1.6e-6]\nTraining steps:   4%|▍         | 9/200 [04:12<1:26:34, 27.20s/it, grad_norm=0.00241, loss=0.114, lr=1.6e-6]\nTraining steps:   4%|▍         | 9/200 [04:12<1:26:34, 27.20s/it, grad_norm=0.0014, loss=0.075, lr=1.8e-6] \nTraining steps:   5%|▌         | 10/200 [04:44<1:30:24, 28.55s/it, grad_norm=0.0014, loss=0.075, lr=1.8e-6]\nTraining steps:   5%|▌         | 10/200 [04:44<1:30:24, 28.55s/it, grad_norm=0.0117, loss=0.138, lr=2e-6]  \nTraining steps:   6%|▌         | 11/200 [05:10<1:27:53, 27.90s/it, grad_norm=0.0117, loss=0.138, lr=2e-6]\nTraining steps:   6%|▌         | 11/200 [05:10<1:27:53, 27.90s/it, grad_norm=0.00325, loss=0.0685, lr=2.2e-6]\nTraining steps:   6%|▌         | 12/200 [05:36<1:25:05, 27.16s/it, grad_norm=0.00325, loss=0.0685, lr=2.2e-6]\nTraining steps:   6%|▌         | 12/200 [05:36<1:25:05, 27.16s/it, grad_norm=0.0304, loss=0.161, lr=2.4e-6]  \nTraining steps:   6%|▋         | 13/200 [06:07<1:28:48, 28.50s/it, grad_norm=0.0304, loss=0.161, lr=2.4e-6]\nTraining steps:   6%|▋         | 13/200 [06:07<1:28:48, 28.50s/it, grad_norm=0.00269, loss=0.109, lr=2.6e-6]\nTraining steps:   7%|▋         | 14/200 [06:34<1:26:25, 27.88s/it, grad_norm=0.00269, loss=0.109, lr=2.6e-6]\nTraining steps:   7%|▋         | 14/200 [06:34<1:26:25, 27.88s/it, grad_norm=0.00189, loss=0.0642, lr=2.8e-6]\nTraining steps:   8%|▊         | 15/200 [06:59<1:23:41, 27.15s/it, grad_norm=0.00189, loss=0.0642, lr=2.8e-6]\nTraining steps:   8%|▊         | 15/200 [06:59<1:23:41, 27.15s/it, grad_norm=0.0013, loss=0.0592, lr=3e-6]   \nTraining steps:   8%|▊         | 16/200 [07:31<1:27:20, 28.48s/it, grad_norm=0.0013, loss=0.0592, lr=3e-6]\nTraining steps:   8%|▊         | 16/200 [07:31<1:27:20, 28.48s/it, grad_norm=0.0378, loss=0.249, lr=3.2e-6]\nTraining steps:   8%|▊         | 17/200 [07:57<1:25:00, 27.87s/it, grad_norm=0.0378, loss=0.249, lr=3.2e-6]\nTraining steps:   8%|▊         | 17/200 [07:57<1:25:00, 27.87s/it, grad_norm=0.149, loss=0.46, lr=3.4e-6]  \nTraining steps:   9%|▉         | 18/200 [08:22<1:22:19, 27.14s/it, grad_norm=0.149, loss=0.46, lr=3.4e-6]\nTraining steps:   9%|▉         | 18/200 [08:22<1:22:19, 27.14s/it, grad_norm=0.00461, loss=0.0528, lr=3.6e-6]\nTraining steps:  10%|▉         | 19/200 [08:54<1:25:51, 28.46s/it, grad_norm=0.00461, loss=0.0528, lr=3.6e-6]\nTraining steps:  10%|▉         | 19/200 [08:54<1:25:51, 28.46s/it, grad_norm=0.00388, loss=0.154, lr=3.8e-6] \nTraining steps:  10%|█         | 20/200 [09:20<1:23:32, 27.84s/it, grad_norm=0.00388, loss=0.154, lr=3.8e-6]\nTraining steps:  10%|█         | 20/200 [09:20<1:23:32, 27.84s/it, grad_norm=0.00138, loss=0.0657, lr=4e-6] \nTraining steps:  10%|█         | 21/200 [09:46<1:20:53, 27.11s/it, grad_norm=0.00138, loss=0.0657, lr=4e-6]\nTraining steps:  10%|█         | 21/200 [09:46<1:20:53, 27.11s/it, grad_norm=0.00409, loss=0.0463, lr=4.2e-6]\nTraining steps:  11%|█         | 22/200 [10:18<1:25:00, 28.66s/it, grad_norm=0.00409, loss=0.0463, lr=4.2e-6]\nTraining steps:  11%|█         | 22/200 [10:18<1:25:00, 28.66s/it, grad_norm=0.00372, loss=0.159, lr=4.4e-6] \nTraining steps:  12%|█▏        | 23/200 [10:45<1:22:36, 28.01s/it, grad_norm=0.00372, loss=0.159, lr=4.4e-6]\nTraining steps:  12%|█▏        | 23/200 [10:45<1:22:36, 28.01s/it, grad_norm=0.00233, loss=0.17, lr=4.6e-6] \nTraining steps:  12%|█▏        | 24/200 [11:10<1:19:56, 27.25s/it, grad_norm=0.00233, loss=0.17, lr=4.6e-6]\nTraining steps:  12%|█▏        | 24/200 [11:10<1:19:56, 27.25s/it, grad_norm=0.00132, loss=0.0647, lr=4.8e-6]\nTraining steps:  12%|█▎        | 25/200 [11:44<1:25:17, 29.24s/it, grad_norm=0.00132, loss=0.0647, lr=4.8e-6]\nTraining steps:  12%|█▎        | 25/200 [11:44<1:25:17, 29.24s/it, grad_norm=0.00305, loss=0.0967, lr=5e-6]  \nTraining steps:  13%|█▎        | 26/200 [12:10<1:22:22, 28.41s/it, grad_norm=0.00305, loss=0.0967, lr=5e-6]\nTraining steps:  13%|█▎        | 26/200 [12:10<1:22:22, 28.41s/it, grad_norm=0.0016, loss=0.0744, lr=5.2e-6]\nTraining steps:  14%|█▎        | 27/200 [12:36<1:19:19, 27.51s/it, grad_norm=0.0016, loss=0.0744, lr=5.2e-6]\nTraining steps:  14%|█▎        | 27/200 [12:36<1:19:19, 27.51s/it, grad_norm=0.00127, loss=0.064, lr=5.4e-6]\nTraining steps:  14%|█▍        | 28/200 [13:07<1:22:21, 28.73s/it, grad_norm=0.00127, loss=0.064, lr=5.4e-6]\nTraining steps:  14%|█▍        | 28/200 [13:07<1:22:21, 28.73s/it, grad_norm=0.00427, loss=0.215, lr=5.6e-6]\nTraining steps:  14%|█▍        | 29/200 [13:34<1:19:56, 28.05s/it, grad_norm=0.00427, loss=0.215, lr=5.6e-6]\nTraining steps:  14%|█▍        | 29/200 [13:34<1:19:56, 28.05s/it, grad_norm=0.00182, loss=0.0676, lr=5.8e-6]\nTraining steps:  15%|█▌        | 30/200 [13:59<1:17:14, 27.26s/it, grad_norm=0.00182, loss=0.0676, lr=5.8e-6]\nTraining steps:  15%|█▌        | 30/200 [13:59<1:17:14, 27.26s/it, grad_norm=0.00201, loss=0.0361, lr=6e-6]  \nTraining steps:  16%|█▌        | 31/200 [14:31<1:20:32, 28.59s/it, grad_norm=0.00201, loss=0.0361, lr=6e-6]\nTraining steps:  16%|█▌        | 31/200 [14:31<1:20:32, 28.59s/it, grad_norm=0.00275, loss=0.106, lr=6.2e-6]\nTraining steps:  16%|█▌        | 32/200 [14:57<1:18:16, 27.95s/it, grad_norm=0.00275, loss=0.106, lr=6.2e-6]\nTraining steps:  16%|█▌        | 32/200 [14:57<1:18:16, 27.95s/it, grad_norm=0.00229, loss=0.165, lr=6.4e-6]\nTraining steps:  16%|█▋        | 33/200 [15:23<1:15:42, 27.20s/it, grad_norm=0.00229, loss=0.165, lr=6.4e-6]\nTraining steps:  16%|█▋        | 33/200 [15:23<1:15:42, 27.20s/it, grad_norm=0.00323, loss=0.0389, lr=6.6e-6]\nTraining steps:  17%|█▋        | 34/200 [15:55<1:18:56, 28.53s/it, grad_norm=0.00323, loss=0.0389, lr=6.6e-6]\nTraining steps:  17%|█▋        | 34/200 [15:55<1:18:56, 28.53s/it, grad_norm=0.00333, loss=0.095, lr=6.8e-6] \nTraining steps:  18%|█▊        | 35/200 [16:21<1:16:45, 27.91s/it, grad_norm=0.00333, loss=0.095, lr=6.8e-6]\nTraining steps:  18%|█▊        | 35/200 [16:21<1:16:45, 27.91s/it, grad_norm=0.00232, loss=0.136, lr=7e-6]  \nTraining steps:  18%|█▊        | 36/200 [16:46<1:14:17, 27.18s/it, grad_norm=0.00232, loss=0.136, lr=7e-6]\nTraining steps:  18%|█▊        | 36/200 [16:46<1:14:17, 27.18s/it, grad_norm=0.00132, loss=0.0457, lr=7.2e-6]\nTraining steps:  18%|█▊        | 37/200 [17:18<1:17:27, 28.51s/it, grad_norm=0.00132, loss=0.0457, lr=7.2e-6]\nTraining steps:  18%|█▊        | 37/200 [17:18<1:17:27, 28.51s/it, grad_norm=0.00446, loss=0.103, lr=7.4e-6] \nTraining steps:  19%|█▉        | 38/200 [17:45<1:15:19, 27.90s/it, grad_norm=0.00446, loss=0.103, lr=7.4e-6]\nTraining steps:  19%|█▉        | 38/200 [17:45<1:15:19, 27.90s/it, grad_norm=0.0791, loss=0.371, lr=7.6e-6] \nTraining steps:  20%|█▉        | 39/200 [18:10<1:12:52, 27.16s/it, grad_norm=0.0791, loss=0.371, lr=7.6e-6]\nTraining steps:  20%|█▉        | 39/200 [18:10<1:12:52, 27.16s/it, grad_norm=0.00497, loss=0.335, lr=7.8e-6]\nTraining steps:  20%|██        | 40/200 [18:42<1:16:04, 28.53s/it, grad_norm=0.00497, loss=0.335, lr=7.8e-6]\nTraining steps:  20%|██        | 40/200 [18:42<1:16:04, 28.53s/it, grad_norm=0.00549, loss=0.109, lr=8e-6]  \nTraining steps:  20%|██        | 41/200 [19:08<1:13:56, 27.91s/it, grad_norm=0.00549, loss=0.109, lr=8e-6]\nTraining steps:  20%|██        | 41/200 [19:08<1:13:56, 27.91s/it, grad_norm=0.00221, loss=0.11, lr=8.2e-6]\nTraining steps:  21%|██        | 42/200 [19:34<1:11:33, 27.17s/it, grad_norm=0.00221, loss=0.11, lr=8.2e-6]\nTraining steps:  21%|██        | 42/200 [19:34<1:11:33, 27.17s/it, grad_norm=0.0015, loss=0.037, lr=8.4e-6]\nTraining steps:  22%|██▏       | 43/200 [20:05<1:14:43, 28.56s/it, grad_norm=0.0015, loss=0.037, lr=8.4e-6]\nTraining steps:  22%|██▏       | 43/200 [20:05<1:14:43, 28.56s/it, grad_norm=0.00616, loss=0.351, lr=8.6e-6]\nTraining steps:  22%|██▏       | 44/200 [20:32<1:12:37, 27.93s/it, grad_norm=0.00616, loss=0.351, lr=8.6e-6]\nTraining steps:  22%|██▏       | 44/200 [20:32<1:12:37, 27.93s/it, grad_norm=0.013, loss=0.119, lr=8.8e-6]  \nTraining steps:  22%|██▎       | 45/200 [20:57<1:10:13, 27.18s/it, grad_norm=0.013, loss=0.119, lr=8.8e-6]\nTraining steps:  22%|██▎       | 45/200 [20:57<1:10:13, 27.18s/it, grad_norm=0.00723, loss=0.0629, lr=9e-6]\nTraining steps:  23%|██▎       | 46/200 [21:29<1:13:14, 28.53s/it, grad_norm=0.00723, loss=0.0629, lr=9e-6]\nTraining steps:  23%|██▎       | 46/200 [21:29<1:13:14, 28.53s/it, grad_norm=0.0032, loss=0.0942, lr=9.2e-6]\nTraining steps:  24%|██▎       | 47/200 [21:55<1:11:10, 27.91s/it, grad_norm=0.0032, loss=0.0942, lr=9.2e-6]\nTraining steps:  24%|██▎       | 47/200 [21:55<1:11:10, 27.91s/it, grad_norm=0.00149, loss=0.0651, lr=9.4e-6]\nTraining steps:  24%|██▍       | 48/200 [22:21<1:08:49, 27.17s/it, grad_norm=0.00149, loss=0.0651, lr=9.4e-6]\nTraining steps:  24%|██▍       | 48/200 [22:21<1:08:49, 27.17s/it, grad_norm=0.00282, loss=0.22, lr=9.6e-6]  \nTraining steps:  24%|██▍       | 49/200 [22:53<1:11:51, 28.55s/it, grad_norm=0.00282, loss=0.22, lr=9.6e-6]\nTraining steps:  24%|██▍       | 49/200 [22:53<1:11:51, 28.55s/it, grad_norm=0.00293, loss=0.114, lr=9.8e-6]\nTraining steps:  25%|██▌       | 50/200 [23:19<1:09:50, 27.93s/it, grad_norm=0.00293, loss=0.114, lr=9.8e-6]\nTraining steps:  25%|██▌       | 50/200 [23:19<1:09:50, 27.93s/it, grad_norm=0.00169, loss=0.0856, lr=1e-5] \nTraining steps:  26%|██▌       | 51/200 [23:45<1:07:31, 27.19s/it, grad_norm=0.00169, loss=0.0856, lr=1e-5]\nTraining steps:  26%|██▌       | 51/200 [23:45<1:07:31, 27.19s/it, grad_norm=0.0025, loss=0.0367, lr=1.02e-5]\nTraining steps:  26%|██▌       | 52/200 [24:16<1:10:22, 28.53s/it, grad_norm=0.0025, loss=0.0367, lr=1.02e-5]\nTraining steps:  26%|██▌       | 52/200 [24:16<1:10:22, 28.53s/it, grad_norm=0.00562, loss=0.107, lr=1.04e-5]\nTraining steps:  26%|██▋       | 53/200 [24:43<1:08:22, 27.91s/it, grad_norm=0.00562, loss=0.107, lr=1.04e-5]\nTraining steps:  26%|██▋       | 53/200 [24:43<1:08:22, 27.91s/it, grad_norm=0.00313, loss=0.0692, lr=1.06e-5]\nTraining steps:  27%|██▋       | 54/200 [25:08<1:06:07, 27.17s/it, grad_norm=0.00313, loss=0.0692, lr=1.06e-5]\nTraining steps:  27%|██▋       | 54/200 [25:08<1:06:07, 27.17s/it, grad_norm=0.00147, loss=0.0369, lr=1.08e-5]\nTraining steps:  28%|██▊       | 55/200 [25:40<1:09:00, 28.55s/it, grad_norm=0.00147, loss=0.0369, lr=1.08e-5]\nTraining steps:  28%|██▊       | 55/200 [25:40<1:09:00, 28.55s/it, grad_norm=0.00365, loss=0.153, lr=1.1e-5]  \nTraining steps:  28%|██▊       | 56/200 [26:06<1:07:00, 27.92s/it, grad_norm=0.00365, loss=0.153, lr=1.1e-5]\nTraining steps:  28%|██▊       | 56/200 [26:06<1:07:00, 27.92s/it, grad_norm=0.00203, loss=0.0833, lr=1.12e-5]\nTraining steps:  28%|██▊       | 57/200 [26:32<1:04:46, 27.18s/it, grad_norm=0.00203, loss=0.0833, lr=1.12e-5]\nTraining steps:  28%|██▊       | 57/200 [26:32<1:04:46, 27.18s/it, grad_norm=0.00149, loss=0.0398, lr=1.14e-5]\nTraining steps:  29%|██▉       | 58/200 [27:04<1:07:35, 28.56s/it, grad_norm=0.00149, loss=0.0398, lr=1.14e-5]\nTraining steps:  29%|██▉       | 58/200 [27:04<1:07:35, 28.56s/it, grad_norm=0.0117, loss=0.13, lr=1.16e-5]   \nTraining steps:  30%|██▉       | 59/200 [27:30<1:05:38, 27.93s/it, grad_norm=0.0117, loss=0.13, lr=1.16e-5]\nTraining steps:  30%|██▉       | 59/200 [27:30<1:05:38, 27.93s/it, grad_norm=0.00191, loss=0.0865, lr=1.18e-5]\nTraining steps:  30%|███       | 60/200 [27:56<1:03:25, 27.19s/it, grad_norm=0.00191, loss=0.0865, lr=1.18e-5]\nTraining steps:  30%|███       | 60/200 [27:56<1:03:25, 27.19s/it, grad_norm=0.00161, loss=0.0355, lr=1.2e-5] \nTraining steps:  30%|███       | 61/200 [28:27<1:06:03, 28.51s/it, grad_norm=0.00161, loss=0.0355, lr=1.2e-5]\nTraining steps:  30%|███       | 61/200 [28:27<1:06:03, 28.51s/it, grad_norm=0.00328, loss=0.0938, lr=1.22e-5]\nTraining steps:  31%|███       | 62/200 [28:54<1:04:11, 27.91s/it, grad_norm=0.00328, loss=0.0938, lr=1.22e-5]\nTraining steps:  31%|███       | 62/200 [28:54<1:04:11, 27.91s/it, grad_norm=0.00238, loss=0.0908, lr=1.24e-5]\nTraining steps:  32%|███▏      | 63/200 [29:19<1:02:06, 27.20s/it, grad_norm=0.00238, loss=0.0908, lr=1.24e-5]\nTraining steps:  32%|███▏      | 63/200 [29:19<1:02:06, 27.20s/it, grad_norm=0.00131, loss=0.0409, lr=1.26e-5]\nTraining steps:  32%|███▏      | 64/200 [29:51<1:04:40, 28.53s/it, grad_norm=0.00131, loss=0.0409, lr=1.26e-5]\nTraining steps:  32%|███▏      | 64/200 [29:51<1:04:40, 28.53s/it, grad_norm=0.008, loss=0.121, lr=1.28e-5]   \nTraining steps:  32%|███▎      | 65/200 [30:17<1:02:48, 27.92s/it, grad_norm=0.008, loss=0.121, lr=1.28e-5]\nTraining steps:  32%|███▎      | 65/200 [30:17<1:02:48, 27.92s/it, grad_norm=0.00238, loss=0.0648, lr=1.3e-5]\nTraining steps:  33%|███▎      | 66/200 [30:43<1:00:41, 27.18s/it, grad_norm=0.00238, loss=0.0648, lr=1.3e-5]\nTraining steps:  33%|███▎      | 66/200 [30:43<1:00:41, 27.18s/it, grad_norm=0.00221, loss=0.167, lr=1.32e-5]\nTraining steps:  34%|███▎      | 67/200 [31:14<1:03:13, 28.52s/it, grad_norm=0.00221, loss=0.167, lr=1.32e-5]\nTraining steps:  34%|███▎      | 67/200 [31:14<1:03:13, 28.52s/it, grad_norm=0.00276, loss=0.106, lr=1.34e-5]\nTraining steps:  34%|███▍      | 68/200 [31:41<1:01:22, 27.89s/it, grad_norm=0.00276, loss=0.106, lr=1.34e-5]\nTraining steps:  34%|███▍      | 68/200 [31:41<1:01:22, 27.89s/it, grad_norm=0.0101, loss=0.782, lr=1.36e-5] \nTraining steps:  34%|███▍      | 69/200 [32:06<59:18, 27.16s/it, grad_norm=0.0101, loss=0.782, lr=1.36e-5]\nTraining steps:  34%|███▍      | 69/200 [32:06<59:18, 27.16s/it, grad_norm=0.00156, loss=0.0424, lr=1.38e-5]\nTraining steps:  35%|███▌      | 70/200 [32:38<1:01:46, 28.51s/it, grad_norm=0.00156, loss=0.0424, lr=1.38e-5]\nTraining steps:  35%|███▌      | 70/200 [32:38<1:01:46, 28.51s/it, grad_norm=0.00406, loss=0.0986, lr=1.4e-5] \nTraining steps:  36%|███▌      | 71/200 [33:04<59:59, 27.90s/it, grad_norm=0.00406, loss=0.0986, lr=1.4e-5]\nTraining steps:  36%|███▌      | 71/200 [33:04<59:59, 27.90s/it, grad_norm=0.00406, loss=0.074, lr=1.42e-5]\nTraining steps:  36%|███▌      | 72/200 [33:30<57:58, 27.17s/it, grad_norm=0.00406, loss=0.074, lr=1.42e-5]\nTraining steps:  36%|███▌      | 72/200 [33:30<57:58, 27.17s/it, grad_norm=0.0027, loss=0.0375, lr=1.44e-5]\nTraining steps:  36%|███▋      | 73/200 [34:01<1:00:16, 28.48s/it, grad_norm=0.0027, loss=0.0375, lr=1.44e-5]\nTraining steps:  36%|███▋      | 73/200 [34:01<1:00:16, 28.48s/it, grad_norm=0.00269, loss=0.105, lr=1.46e-5]\nTraining steps:  37%|███▋      | 74/200 [34:28<58:30, 27.86s/it, grad_norm=0.00269, loss=0.105, lr=1.46e-5]\nTraining steps:  37%|███▋      | 74/200 [34:28<58:30, 27.86s/it, grad_norm=0.0109, loss=0.803, lr=1.48e-5] \nTraining steps:  38%|███▊      | 75/200 [34:53<56:32, 27.14s/it, grad_norm=0.0109, loss=0.803, lr=1.48e-5]\nTraining steps:  38%|███▊      | 75/200 [34:53<56:32, 27.14s/it, grad_norm=0.0016, loss=0.0474, lr=1.5e-5]\nTraining steps:  38%|███▊      | 76/200 [35:25<59:05, 28.59s/it, grad_norm=0.0016, loss=0.0474, lr=1.5e-5]\nTraining steps:  38%|███▊      | 76/200 [35:25<59:05, 28.59s/it, grad_norm=0.00421, loss=0.0982, lr=1.52e-5]\nTraining steps:  38%|███▊      | 77/200 [35:52<57:19, 27.96s/it, grad_norm=0.00421, loss=0.0982, lr=1.52e-5]\nTraining steps:  38%|███▊      | 77/200 [35:52<57:19, 27.96s/it, grad_norm=0.00357, loss=0.278, lr=1.54e-5] \nTraining steps:  39%|███▉      | 78/200 [36:17<55:19, 27.21s/it, grad_norm=0.00357, loss=0.278, lr=1.54e-5]\nTraining steps:  39%|███▉      | 78/200 [36:17<55:19, 27.21s/it, grad_norm=0.00158, loss=0.105, lr=1.56e-5]\nTraining steps:  40%|███▉      | 79/200 [36:49<57:41, 28.60s/it, grad_norm=0.00158, loss=0.105, lr=1.56e-5]\nTraining steps:  40%|███▉      | 79/200 [36:49<57:41, 28.60s/it, grad_norm=0.00482, loss=0.264, lr=1.58e-5]\nTraining steps:  40%|████      | 80/200 [37:16<55:57, 27.98s/it, grad_norm=0.00482, loss=0.264, lr=1.58e-5]\nTraining steps:  40%|████      | 80/200 [37:16<55:57, 27.98s/it, grad_norm=0.00928, loss=0.876, lr=1.6e-5] \nTraining steps:  40%|████      | 81/200 [37:41<53:58, 27.21s/it, grad_norm=0.00928, loss=0.876, lr=1.6e-5]\nTraining steps:  40%|████      | 81/200 [37:41<53:58, 27.21s/it, grad_norm=0.0223, loss=0.897, lr=1.62e-5]\nTraining steps:  41%|████      | 82/200 [38:13<56:17, 28.63s/it, grad_norm=0.0223, loss=0.897, lr=1.62e-5]\nTraining steps:  41%|████      | 82/200 [38:13<56:17, 28.63s/it, grad_norm=0.00623, loss=0.352, lr=1.64e-5]\nTraining steps:  42%|████▏     | 83/200 [38:39<54:35, 27.99s/it, grad_norm=0.00623, loss=0.352, lr=1.64e-5]\nTraining steps:  42%|████▏     | 83/200 [38:39<54:35, 27.99s/it, grad_norm=0.00934, loss=0.817, lr=1.66e-5]\nTraining steps:  42%|████▏     | 84/200 [39:05<52:40, 27.24s/it, grad_norm=0.00934, loss=0.817, lr=1.66e-5]\nTraining steps:  42%|████▏     | 84/200 [39:05<52:40, 27.24s/it, grad_norm=0.00266, loss=0.197, lr=1.68e-5]\nTraining steps:  42%|████▎     | 85/200 [39:39<56:06, 29.28s/it, grad_norm=0.00266, loss=0.197, lr=1.68e-5]\nTraining steps:  42%|████▎     | 85/200 [39:39<56:06, 29.28s/it, grad_norm=0.0104, loss=0.126, lr=1.7e-5]  \nTraining steps:  43%|████▎     | 86/200 [40:05<54:01, 28.43s/it, grad_norm=0.0104, loss=0.126, lr=1.7e-5]\nTraining steps:  43%|████▎     | 86/200 [40:05<54:01, 28.43s/it, grad_norm=0.00163, loss=0.0636, lr=1.72e-5]\nTraining steps:  44%|████▎     | 87/200 [40:31<51:51, 27.54s/it, grad_norm=0.00163, loss=0.0636, lr=1.72e-5]\nTraining steps:  44%|████▎     | 87/200 [40:31<51:51, 27.54s/it, grad_norm=0.00163, loss=0.127, lr=1.74e-5] \nTraining steps:  44%|████▍     | 88/200 [41:03<53:45, 28.80s/it, grad_norm=0.00163, loss=0.127, lr=1.74e-5]\nTraining steps:  44%|████▍     | 88/200 [41:03<53:45, 28.80s/it, grad_norm=0.00299, loss=0.123, lr=1.76e-5]\nTraining steps:  44%|████▍     | 89/200 [41:29<52:00, 28.11s/it, grad_norm=0.00299, loss=0.123, lr=1.76e-5]\nTraining steps:  44%|████▍     | 89/200 [41:29<52:00, 28.11s/it, grad_norm=0.0022, loss=0.086, lr=1.78e-5] \nTraining steps:  45%|████▌     | 90/200 [41:55<50:05, 27.32s/it, grad_norm=0.0022, loss=0.086, lr=1.78e-5]\nTraining steps:  45%|████▌     | 90/200 [41:55<50:05, 27.32s/it, grad_norm=0.00149, loss=0.0397, lr=1.8e-5]\nTraining steps:  46%|████▌     | 91/200 [42:31<54:32, 30.02s/it, grad_norm=0.00149, loss=0.0397, lr=1.8e-5]\nTraining steps:  46%|████▌     | 91/200 [42:31<54:32, 30.02s/it, grad_norm=0.00255, loss=0.0957, lr=1.82e-5]\nTraining steps:  46%|████▌     | 92/200 [42:57<52:06, 28.95s/it, grad_norm=0.00255, loss=0.0957, lr=1.82e-5]\nTraining steps:  46%|████▌     | 92/200 [42:57<52:06, 28.95s/it, grad_norm=0.00812, loss=0.1, lr=1.84e-5]   \nTraining steps:  46%|████▋     | 93/200 [43:23<49:45, 27.90s/it, grad_norm=0.00812, loss=0.1, lr=1.84e-5]\nTraining steps:  46%|████▋     | 93/200 [43:23<49:45, 27.90s/it, grad_norm=0.00155, loss=0.0369, lr=1.86e-5]\nTraining steps:  47%|████▋     | 94/200 [43:55<51:16, 29.02s/it, grad_norm=0.00155, loss=0.0369, lr=1.86e-5]\nTraining steps:  47%|████▋     | 94/200 [43:55<51:16, 29.02s/it, grad_norm=0.00349, loss=0.0959, lr=1.88e-5]\nTraining steps:  48%|████▊     | 95/200 [44:21<49:26, 28.25s/it, grad_norm=0.00349, loss=0.0959, lr=1.88e-5]\nTraining steps:  48%|████▊     | 95/200 [44:21<49:26, 28.25s/it, grad_norm=0.00149, loss=0.0666, lr=1.9e-5] \nTraining steps:  48%|████▊     | 96/200 [44:46<47:31, 27.42s/it, grad_norm=0.00149, loss=0.0666, lr=1.9e-5]\nTraining steps:  48%|████▊     | 96/200 [44:46<47:31, 27.42s/it, grad_norm=0.00125, loss=0.0551, lr=1.92e-5]\nTraining steps:  48%|████▊     | 97/200 [45:18<49:16, 28.70s/it, grad_norm=0.00125, loss=0.0551, lr=1.92e-5]\nTraining steps:  48%|████▊     | 97/200 [45:18<49:16, 28.70s/it, grad_norm=0.00708, loss=0.114, lr=1.94e-5] \nTraining steps:  49%|████▉     | 98/200 [45:45<47:39, 28.03s/it, grad_norm=0.00708, loss=0.114, lr=1.94e-5]\nTraining steps:  49%|████▉     | 98/200 [45:45<47:39, 28.03s/it, grad_norm=0.0171, loss=0.115, lr=1.96e-5] \nTraining steps:  50%|████▉     | 99/200 [46:10<45:53, 27.26s/it, grad_norm=0.0171, loss=0.115, lr=1.96e-5]\nTraining steps:  50%|████▉     | 99/200 [46:10<45:53, 27.26s/it, grad_norm=0.00125, loss=0.0533, lr=1.98e-5]\nTraining steps:  50%|█████     | 100/200 [46:42<47:35, 28.56s/it, grad_norm=0.00125, loss=0.0533, lr=1.98e-5]\nTraining steps:  50%|█████     | 100/200 [46:42<47:35, 28.56s/it, grad_norm=0.00427, loss=0.208, lr=2e-5]    \nTraining steps:  50%|█████     | 101/200 [47:09<46:27, 28.16s/it, grad_norm=0.00427, loss=0.208, lr=2e-5]\nTraining steps:  50%|█████     | 101/200 [47:09<46:27, 28.16s/it, grad_norm=0.641, loss=0.958, lr=2e-5]  \nTraining steps:  51%|█████     | 102/200 [47:34<44:40, 27.35s/it, grad_norm=0.641, loss=0.958, lr=2e-5]\nTraining steps:  51%|█████     | 102/200 [47:34<44:40, 27.35s/it, grad_norm=0.00124, loss=0.0521, lr=2e-5]\nTraining steps:  52%|█████▏    | 103/200 [48:06<46:25, 28.72s/it, grad_norm=0.00124, loss=0.0521, lr=2e-5]\nTraining steps:  52%|█████▏    | 103/200 [48:06<46:25, 28.72s/it, grad_norm=0.0097, loss=0.127, lr=2e-5]  \nTraining steps:  52%|█████▏    | 104/200 [48:33<44:53, 28.06s/it, grad_norm=0.0097, loss=0.127, lr=2e-5]\nTraining steps:  52%|█████▏    | 104/200 [48:33<44:53, 28.06s/it, grad_norm=0.00352, loss=0.0675, lr=2e-5]\nTraining steps:  52%|█████▎    | 105/200 [48:58<43:12, 27.29s/it, grad_norm=0.00352, loss=0.0675, lr=2e-5]\nTraining steps:  52%|█████▎    | 105/200 [48:58<43:12, 27.29s/it, grad_norm=0.00848, loss=0.0672, lr=2e-5]\nTraining steps:  53%|█████▎    | 106/200 [49:30<44:53, 28.66s/it, grad_norm=0.00848, loss=0.0672, lr=2e-5]\nTraining steps:  53%|█████▎    | 106/200 [49:30<44:53, 28.66s/it, grad_norm=0.0182, loss=0.161, lr=2e-5]  \nTraining steps:  54%|█████▎    | 107/200 [49:57<43:25, 28.01s/it, grad_norm=0.0182, loss=0.161, lr=2e-5]\nTraining steps:  54%|█████▎    | 107/200 [49:57<43:25, 28.01s/it, grad_norm=0.00227, loss=0.096, lr=2e-5]\nTraining steps:  54%|█████▍    | 108/200 [50:22<41:46, 27.24s/it, grad_norm=0.00227, loss=0.096, lr=2e-5]\nTraining steps:  54%|█████▍    | 108/200 [50:22<41:46, 27.24s/it, grad_norm=0.00266, loss=0.199, lr=2e-5]\nTraining steps:  55%|█████▍    | 109/200 [50:54<43:29, 28.67s/it, grad_norm=0.00266, loss=0.199, lr=2e-5]\nTraining steps:  55%|█████▍    | 109/200 [50:54<43:29, 28.67s/it, grad_norm=0.0043, loss=0.0973, lr=2e-5]\nTraining steps:  55%|█████▌    | 110/200 [51:21<42:02, 28.03s/it, grad_norm=0.0043, loss=0.0973, lr=2e-5]\nTraining steps:  55%|█████▌    | 110/200 [51:21<42:02, 28.03s/it, grad_norm=0.00209, loss=0.0753, lr=2e-5]\nTraining steps:  56%|█████▌    | 111/200 [51:46<40:26, 27.27s/it, grad_norm=0.00209, loss=0.0753, lr=2e-5]\nTraining steps:  56%|█████▌    | 111/200 [51:46<40:26, 27.27s/it, grad_norm=0.0017, loss=0.0346, lr=2e-5] \nTraining steps:  56%|█████▌    | 112/200 [52:18<41:53, 28.56s/it, grad_norm=0.0017, loss=0.0346, lr=2e-5]\nTraining steps:  56%|█████▌    | 112/200 [52:18<41:53, 28.56s/it, grad_norm=0.00255, loss=0.101, lr=2e-5]\nTraining steps:  56%|█████▋    | 113/200 [52:44<40:29, 27.93s/it, grad_norm=0.00255, loss=0.101, lr=2e-5]\nTraining steps:  56%|█████▋    | 113/200 [52:44<40:29, 27.93s/it, grad_norm=0.00212, loss=0.105, lr=2e-5]\nTraining steps:  57%|█████▋    | 114/200 [53:10<38:58, 27.19s/it, grad_norm=0.00212, loss=0.105, lr=2e-5]\nTraining steps:  57%|█████▋    | 114/200 [53:10<38:58, 27.19s/it, grad_norm=0.0203, loss=0.125, lr=2e-5] \nTraining steps:  57%|█████▊    | 115/200 [53:41<40:23, 28.51s/it, grad_norm=0.0203, loss=0.125, lr=2e-5]\nTraining steps:  57%|█████▊    | 115/200 [53:41<40:23, 28.51s/it, grad_norm=0.00333, loss=0.16, lr=2e-5]\nTraining steps:  58%|█████▊    | 116/200 [54:08<39:03, 27.90s/it, grad_norm=0.00333, loss=0.16, lr=2e-5]\nTraining steps:  58%|█████▊    | 116/200 [54:08<39:03, 27.90s/it, grad_norm=0.00964, loss=0.714, lr=2e-5]\nTraining steps:  58%|█████▊    | 117/200 [54:33<37:35, 27.17s/it, grad_norm=0.00964, loss=0.714, lr=2e-5]\nTraining steps:  58%|█████▊    | 117/200 [54:33<37:35, 27.17s/it, grad_norm=0.00156, loss=0.108, lr=2e-5]\nTraining steps:  59%|█████▉    | 118/200 [55:05<38:58, 28.52s/it, grad_norm=0.00156, loss=0.108, lr=2e-5]\nTraining steps:  59%|█████▉    | 118/200 [55:05<38:58, 28.52s/it, grad_norm=0.00485, loss=0.259, lr=2e-5]\nTraining steps:  60%|█████▉    | 119/200 [55:31<37:40, 27.91s/it, grad_norm=0.00485, loss=0.259, lr=2e-5]\nTraining steps:  60%|█████▉    | 119/200 [55:31<37:40, 27.91s/it, grad_norm=0.00159, loss=0.0708, lr=2e-5]\nTraining steps:  60%|██████    | 120/200 [55:57<36:14, 27.18s/it, grad_norm=0.00159, loss=0.0708, lr=2e-5]\nTraining steps:  60%|██████    | 120/200 [55:57<36:14, 27.18s/it, grad_norm=0.00778, loss=0.0618, lr=2e-5]\nTraining steps:  60%|██████    | 121/200 [56:28<37:31, 28.51s/it, grad_norm=0.00778, loss=0.0618, lr=2e-5]\nTraining steps:  60%|██████    | 121/200 [56:28<37:31, 28.51s/it, grad_norm=0.00273, loss=0.138, lr=2e-5] \nTraining steps:  61%|██████    | 122/200 [56:55<36:16, 27.90s/it, grad_norm=0.00273, loss=0.138, lr=2e-5]\nTraining steps:  61%|██████    | 122/200 [56:55<36:16, 27.90s/it, grad_norm=0.00806, loss=0.659, lr=2e-5]\nTraining steps:  62%|██████▏   | 123/200 [57:20<34:52, 27.17s/it, grad_norm=0.00806, loss=0.659, lr=2e-5]\nTraining steps:  62%|██████▏   | 123/200 [57:20<34:52, 27.17s/it, grad_norm=0.00166, loss=0.0638, lr=2e-5]\nTraining steps:  62%|██████▏   | 124/200 [57:52<36:04, 28.48s/it, grad_norm=0.00166, loss=0.0638, lr=2e-5]\nTraining steps:  62%|██████▏   | 124/200 [57:52<36:04, 28.48s/it, grad_norm=0.00264, loss=0.124, lr=2e-5] \nTraining steps:  62%|██████▎   | 125/200 [58:18<34:49, 27.86s/it, grad_norm=0.00264, loss=0.124, lr=2e-5]\nTraining steps:  62%|██████▎   | 125/200 [58:18<34:49, 27.86s/it, grad_norm=0.00218, loss=0.0621, lr=2e-5]\nTraining steps:  63%|██████▎   | 126/200 [58:44<33:27, 27.13s/it, grad_norm=0.00218, loss=0.0621, lr=2e-5]\nTraining steps:  63%|██████▎   | 126/200 [58:44<33:27, 27.13s/it, grad_norm=0.00179, loss=0.0342, lr=2e-5]\nTraining steps:  64%|██████▎   | 127/200 [59:16<34:51, 28.65s/it, grad_norm=0.00179, loss=0.0342, lr=2e-5]\nTraining steps:  64%|██████▎   | 127/200 [59:16<34:51, 28.65s/it, grad_norm=0.00235, loss=0.103, lr=2e-5] \nTraining steps:  64%|██████▍   | 128/200 [59:42<33:37, 28.01s/it, grad_norm=0.00235, loss=0.103, lr=2e-5]\nTraining steps:  64%|██████▍   | 128/200 [59:42<33:37, 28.01s/it, grad_norm=0.00418, loss=0.29, lr=2e-5] \nTraining steps:  64%|██████▍   | 129/200 [1:00:08<32:13, 27.24s/it, grad_norm=0.00418, loss=0.29, lr=2e-5]\nTraining steps:  64%|██████▍   | 129/200 [1:00:08<32:13, 27.24s/it, grad_norm=0.00201, loss=0.0354, lr=2e-5]\nTraining steps:  65%|██████▌   | 130/200 [1:00:40<33:25, 28.65s/it, grad_norm=0.00201, loss=0.0354, lr=2e-5]\nTraining steps:  65%|██████▌   | 130/200 [1:00:40<33:25, 28.65s/it, grad_norm=0.00247, loss=0.0971, lr=2e-5]\nTraining steps:  66%|██████▌   | 131/200 [1:01:06<32:12, 28.01s/it, grad_norm=0.00247, loss=0.0971, lr=2e-5]\nTraining steps:  66%|██████▌   | 131/200 [1:01:06<32:12, 28.01s/it, grad_norm=0.0016, loss=0.0643, lr=2e-5] \nTraining steps:  66%|██████▌   | 132/200 [1:01:32<30:52, 27.24s/it, grad_norm=0.0016, loss=0.0643, lr=2e-5]\nTraining steps:  66%|██████▌   | 132/200 [1:01:32<30:52, 27.24s/it, grad_norm=0.0014, loss=0.0827, lr=2e-5]\nTraining steps:  66%|██████▋   | 133/200 [1:02:04<31:58, 28.64s/it, grad_norm=0.0014, loss=0.0827, lr=2e-5]\nTraining steps:  66%|██████▋   | 133/200 [1:02:04<31:58, 28.64s/it, grad_norm=0.00439, loss=0.215, lr=2e-5]\nTraining steps:  67%|██████▋   | 134/200 [1:02:30<30:47, 28.00s/it, grad_norm=0.00439, loss=0.215, lr=2e-5]\nTraining steps:  67%|██████▋   | 134/200 [1:02:30<30:47, 28.00s/it, grad_norm=0.00336, loss=0.0652, lr=2e-5]\nTraining steps:  68%|██████▊   | 135/200 [1:02:56<29:30, 27.23s/it, grad_norm=0.00336, loss=0.0652, lr=2e-5]\nTraining steps:  68%|██████▊   | 135/200 [1:02:56<29:30, 27.23s/it, grad_norm=0.00188, loss=0.0348, lr=2e-5]\nTraining steps:  68%|██████▊   | 136/200 [1:03:27<30:32, 28.63s/it, grad_norm=0.00188, loss=0.0348, lr=2e-5]\nTraining steps:  68%|██████▊   | 136/200 [1:03:27<30:32, 28.63s/it, grad_norm=0.00359, loss=0.189, lr=2e-5] \nTraining steps:  68%|██████▊   | 137/200 [1:03:54<29:22, 27.98s/it, grad_norm=0.00359, loss=0.189, lr=2e-5]\nTraining steps:  68%|██████▊   | 137/200 [1:03:54<29:22, 27.98s/it, grad_norm=0.00188, loss=0.0662, lr=2e-5]\nTraining steps:  69%|██████▉   | 138/200 [1:04:19<28:07, 27.22s/it, grad_norm=0.00188, loss=0.0662, lr=2e-5]\nTraining steps:  69%|██████▉   | 138/200 [1:04:19<28:07, 27.22s/it, grad_norm=0.00284, loss=0.0393, lr=2e-5]\nTraining steps:  70%|██████▉   | 139/200 [1:04:51<29:08, 28.66s/it, grad_norm=0.00284, loss=0.0393, lr=2e-5]\nTraining steps:  70%|██████▉   | 139/200 [1:04:51<29:08, 28.66s/it, grad_norm=0.00836, loss=0.432, lr=2e-5] \nTraining steps:  70%|███████   | 140/200 [1:05:18<28:03, 28.06s/it, grad_norm=0.00836, loss=0.432, lr=2e-5]\nTraining steps:  70%|███████   | 140/200 [1:05:18<28:03, 28.06s/it, grad_norm=0.00214, loss=0.126, lr=2e-5]\nTraining steps:  70%|███████   | 141/200 [1:05:43<26:48, 27.27s/it, grad_norm=0.00214, loss=0.126, lr=2e-5]\nTraining steps:  70%|███████   | 141/200 [1:05:43<26:48, 27.27s/it, grad_norm=0.00156, loss=0.0498, lr=2e-5]\nTraining steps:  71%|███████   | 142/200 [1:06:15<27:41, 28.64s/it, grad_norm=0.00156, loss=0.0498, lr=2e-5]\nTraining steps:  71%|███████   | 142/200 [1:06:15<27:41, 28.64s/it, grad_norm=0.00279, loss=0.146, lr=2e-5] \nTraining steps:  72%|███████▏  | 143/200 [1:06:42<26:35, 27.99s/it, grad_norm=0.00279, loss=0.146, lr=2e-5]\nTraining steps:  72%|███████▏  | 143/200 [1:06:42<26:35, 27.99s/it, grad_norm=0.00302, loss=0.201, lr=2e-5]\nTraining steps:  72%|███████▏  | 144/200 [1:07:07<25:24, 27.23s/it, grad_norm=0.00302, loss=0.201, lr=2e-5]\nTraining steps:  72%|███████▏  | 144/200 [1:07:07<25:24, 27.23s/it, grad_norm=0.00135, loss=0.0374, lr=2e-5]\nTraining steps:  72%|███████▎  | 145/200 [1:07:39<26:15, 28.65s/it, grad_norm=0.00135, loss=0.0374, lr=2e-5]\nTraining steps:  72%|███████▎  | 145/200 [1:07:39<26:15, 28.65s/it, grad_norm=0.00249, loss=0.0983, lr=2e-5]\nTraining steps:  73%|███████▎  | 146/200 [1:08:06<25:11, 27.99s/it, grad_norm=0.00249, loss=0.0983, lr=2e-5]\nTraining steps:  73%|███████▎  | 146/200 [1:08:06<25:11, 27.99s/it, grad_norm=0.00189, loss=0.0706, lr=2e-5]\nTraining steps:  74%|███████▎  | 147/200 [1:08:31<24:02, 27.22s/it, grad_norm=0.00189, loss=0.0706, lr=2e-5]\nTraining steps:  74%|███████▎  | 147/200 [1:08:31<24:02, 27.22s/it, grad_norm=0.00138, loss=0.0396, lr=2e-5]\nTraining steps:  74%|███████▍  | 148/200 [1:09:03<24:46, 28.59s/it, grad_norm=0.00138, loss=0.0396, lr=2e-5]\nTraining steps:  74%|███████▍  | 148/200 [1:09:03<24:46, 28.59s/it, grad_norm=0.0083, loss=0.112, lr=2e-5]  \nTraining steps:  74%|███████▍  | 149/200 [1:09:29<23:45, 27.94s/it, grad_norm=0.0083, loss=0.112, lr=2e-5]\nTraining steps:  74%|███████▍  | 149/200 [1:09:29<23:45, 27.94s/it, grad_norm=0.00186, loss=0.11, lr=2e-5]\nTraining steps:  75%|███████▌  | 150/200 [1:09:55<22:41, 27.22s/it, grad_norm=0.00186, loss=0.11, lr=2e-5]\nTraining steps:  75%|███████▌  | 150/200 [1:09:55<22:41, 27.22s/it, grad_norm=0.314, loss=0.753, lr=2e-5] \nTraining steps:  76%|███████▌  | 151/200 [1:10:27<23:19, 28.56s/it, grad_norm=0.314, loss=0.753, lr=2e-5]\nTraining steps:  76%|███████▌  | 151/200 [1:10:27<23:19, 28.56s/it, grad_norm=0.00299, loss=0.163, lr=2e-5]\nTraining steps:  76%|███████▌  | 152/200 [1:10:53<22:20, 27.93s/it, grad_norm=0.00299, loss=0.163, lr=2e-5]\nTraining steps:  76%|███████▌  | 152/200 [1:10:53<22:20, 27.93s/it, grad_norm=0.00233, loss=0.0706, lr=2e-5]\nTraining steps:  76%|███████▋  | 153/200 [1:11:18<21:17, 27.18s/it, grad_norm=0.00233, loss=0.0706, lr=2e-5]\nTraining steps:  76%|███████▋  | 153/200 [1:11:18<21:17, 27.18s/it, grad_norm=0.00223, loss=0.0352, lr=2e-5]\nTraining steps:  77%|███████▋  | 154/200 [1:11:50<21:52, 28.53s/it, grad_norm=0.00223, loss=0.0352, lr=2e-5]\nTraining steps:  77%|███████▋  | 154/200 [1:11:50<21:52, 28.53s/it, grad_norm=0.00922, loss=0.503, lr=2e-5] \nTraining steps:  78%|███████▊  | 155/200 [1:12:17<20:55, 27.91s/it, grad_norm=0.00922, loss=0.503, lr=2e-5]\nTraining steps:  78%|███████▊  | 155/200 [1:12:17<20:55, 27.91s/it, grad_norm=0.00157, loss=0.062, lr=2e-5]\nTraining steps:  78%|███████▊  | 156/200 [1:12:42<19:55, 27.17s/it, grad_norm=0.00157, loss=0.062, lr=2e-5]\nTraining steps:  78%|███████▊  | 156/200 [1:12:42<19:55, 27.17s/it, grad_norm=0.00952, loss=0.0703, lr=2e-5]\nTraining steps:  78%|███████▊  | 157/200 [1:13:14<20:25, 28.50s/it, grad_norm=0.00952, loss=0.0703, lr=2e-5]\nTraining steps:  78%|███████▊  | 157/200 [1:13:14<20:25, 28.50s/it, grad_norm=0.0581, loss=0.322, lr=2e-5]  \nTraining steps:  79%|███████▉  | 158/200 [1:13:40<19:31, 27.88s/it, grad_norm=0.0581, loss=0.322, lr=2e-5]\nTraining steps:  79%|███████▉  | 158/200 [1:13:40<19:31, 27.88s/it, grad_norm=0.00574, loss=0.0763, lr=2e-5]\nTraining steps:  80%|███████▉  | 159/200 [1:14:05<18:33, 27.15s/it, grad_norm=0.00574, loss=0.0763, lr=2e-5]\nTraining steps:  80%|███████▉  | 159/200 [1:14:05<18:33, 27.15s/it, grad_norm=0.00139, loss=0.0451, lr=2e-5]\nTraining steps:  80%|████████  | 160/200 [1:14:37<19:01, 28.53s/it, grad_norm=0.00139, loss=0.0451, lr=2e-5]\nTraining steps:  80%|████████  | 160/200 [1:14:37<19:01, 28.53s/it, grad_norm=0.00336, loss=0.0932, lr=2e-5]\nTraining steps:  80%|████████  | 161/200 [1:15:04<18:08, 27.91s/it, grad_norm=0.00336, loss=0.0932, lr=2e-5]\nTraining steps:  80%|████████  | 161/200 [1:15:04<18:08, 27.91s/it, grad_norm=0.00568, loss=0.543, lr=2e-5] \nTraining steps:  81%|████████  | 162/200 [1:15:29<17:12, 27.16s/it, grad_norm=0.00568, loss=0.543, lr=2e-5]\nTraining steps:  81%|████████  | 162/200 [1:15:29<17:12, 27.16s/it, grad_norm=0.0015, loss=0.0401, lr=2e-5]\nTraining steps:  82%|████████▏ | 163/200 [1:16:01<17:35, 28.53s/it, grad_norm=0.0015, loss=0.0401, lr=2e-5]\nTraining steps:  82%|████████▏ | 163/200 [1:16:01<17:35, 28.53s/it, grad_norm=0.0121, loss=0.855, lr=2e-5] \nTraining steps:  82%|████████▏ | 164/200 [1:16:27<16:44, 27.91s/it, grad_norm=0.0121, loss=0.855, lr=2e-5]\nTraining steps:  82%|████████▏ | 164/200 [1:16:27<16:44, 27.91s/it, grad_norm=0.00696, loss=0.648, lr=2e-5]\nTraining steps:  82%|████████▎ | 165/200 [1:16:53<15:50, 27.17s/it, grad_norm=0.00696, loss=0.648, lr=2e-5]\nTraining steps:  82%|████████▎ | 165/200 [1:16:53<15:50, 27.17s/it, grad_norm=0.00418, loss=0.317, lr=2e-5]\nTraining steps:  83%|████████▎ | 166/200 [1:17:24<16:10, 28.55s/it, grad_norm=0.00418, loss=0.317, lr=2e-5]\nTraining steps:  83%|████████▎ | 166/200 [1:17:24<16:10, 28.55s/it, grad_norm=0.00787, loss=0.43, lr=2e-5] \nTraining steps:  84%|████████▎ | 167/200 [1:17:51<15:21, 27.92s/it, grad_norm=0.00787, loss=0.43, lr=2e-5]\nTraining steps:  84%|████████▎ | 167/200 [1:17:51<15:21, 27.92s/it, grad_norm=0.00769, loss=0.873, lr=2e-5]\nTraining steps:  84%|████████▍ | 168/200 [1:18:16<14:29, 27.17s/it, grad_norm=0.00769, loss=0.873, lr=2e-5]\nTraining steps:  84%|████████▍ | 168/200 [1:18:16<14:29, 27.17s/it, grad_norm=0.0016, loss=0.115, lr=2e-5] \nTraining steps:  84%|████████▍ | 169/200 [1:18:48<14:43, 28.51s/it, grad_norm=0.0016, loss=0.115, lr=2e-5]\nTraining steps:  84%|████████▍ | 169/200 [1:18:48<14:43, 28.51s/it, grad_norm=0.00227, loss=0.12, lr=2e-5]\nTraining steps:  85%|████████▌ | 170/200 [1:19:14<13:57, 27.90s/it, grad_norm=0.00227, loss=0.12, lr=2e-5]\nTraining steps:  85%|████████▌ | 170/200 [1:19:14<13:57, 27.90s/it, grad_norm=0.00479, loss=0.45, lr=2e-5]\nTraining steps:  86%|████████▌ | 171/200 [1:19:40<13:07, 27.16s/it, grad_norm=0.00479, loss=0.45, lr=2e-5]\nTraining steps:  86%|████████▌ | 171/200 [1:19:40<13:07, 27.16s/it, grad_norm=0.00133, loss=0.0492, lr=2e-5]\nTraining steps:  86%|████████▌ | 172/200 [1:20:12<13:18, 28.51s/it, grad_norm=0.00133, loss=0.0492, lr=2e-5]\nTraining steps:  86%|████████▌ | 172/200 [1:20:12<13:18, 28.51s/it, grad_norm=0.00504, loss=0.293, lr=2e-5] \nTraining steps:  86%|████████▋ | 173/200 [1:20:38<12:33, 27.90s/it, grad_norm=0.00504, loss=0.293, lr=2e-5]\nTraining steps:  86%|████████▋ | 173/200 [1:20:38<12:33, 27.90s/it, grad_norm=0.00349, loss=0.289, lr=2e-5]\nTraining steps:  87%|████████▋ | 174/200 [1:21:04<11:46, 27.16s/it, grad_norm=0.00349, loss=0.289, lr=2e-5]\nTraining steps:  87%|████████▋ | 174/200 [1:21:04<11:46, 27.16s/it, grad_norm=0.00272, loss=0.0367, lr=2e-5]\nTraining steps:  88%|████████▊ | 175/200 [1:21:35<11:53, 28.53s/it, grad_norm=0.00272, loss=0.0367, lr=2e-5]\nTraining steps:  88%|████████▊ | 175/200 [1:21:35<11:53, 28.53s/it, grad_norm=0.00249, loss=0.14, lr=2e-5]  \nTraining steps:  88%|████████▊ | 176/200 [1:22:02<11:09, 27.92s/it, grad_norm=0.00249, loss=0.14, lr=2e-5]\nTraining steps:  88%|████████▊ | 176/200 [1:22:02<11:09, 27.92s/it, grad_norm=0.00221, loss=0.063, lr=2e-5]\nTraining steps:  88%|████████▊ | 177/200 [1:22:27<10:24, 27.17s/it, grad_norm=0.00221, loss=0.063, lr=2e-5]\nTraining steps:  88%|████████▊ | 177/200 [1:22:27<10:24, 27.17s/it, grad_norm=0.00166, loss=0.0366, lr=2e-5]\nTraining steps:  89%|████████▉ | 178/200 [1:22:59<10:27, 28.52s/it, grad_norm=0.00166, loss=0.0366, lr=2e-5]\nTraining steps:  89%|████████▉ | 178/200 [1:22:59<10:27, 28.52s/it, grad_norm=0.00226, loss=0.0994, lr=2e-5]\nTraining steps:  90%|████████▉ | 179/200 [1:23:25<09:46, 27.91s/it, grad_norm=0.00226, loss=0.0994, lr=2e-5]\nTraining steps:  90%|████████▉ | 179/200 [1:23:25<09:46, 27.91s/it, grad_norm=0.00279, loss=0.18, lr=2e-5]  \nTraining steps:  90%|█████████ | 180/200 [1:23:51<09:03, 27.17s/it, grad_norm=0.00279, loss=0.18, lr=2e-5]\nTraining steps:  90%|█████████ | 180/200 [1:23:51<09:03, 27.17s/it, grad_norm=0.00135, loss=0.0707, lr=2e-5]\nTraining steps:  90%|█████████ | 181/200 [1:24:22<09:02, 28.53s/it, grad_norm=0.00135, loss=0.0707, lr=2e-5]\nTraining steps:  90%|█████████ | 181/200 [1:24:22<09:02, 28.53s/it, grad_norm=0.0061, loss=0.107, lr=2e-5]  \nTraining steps:  91%|█████████ | 182/200 [1:24:49<08:22, 27.91s/it, grad_norm=0.0061, loss=0.107, lr=2e-5]\nTraining steps:  91%|█████████ | 182/200 [1:24:49<08:22, 27.91s/it, grad_norm=0.00235, loss=0.0644, lr=2e-5]\nTraining steps:  92%|█████████▏| 183/200 [1:25:14<07:41, 27.17s/it, grad_norm=0.00235, loss=0.0644, lr=2e-5]\nTraining steps:  92%|█████████▏| 183/200 [1:25:14<07:41, 27.17s/it, grad_norm=0.00109, loss=0.0565, lr=2e-5]\nTraining steps:  92%|█████████▏| 184/200 [1:25:46<07:36, 28.56s/it, grad_norm=0.00109, loss=0.0565, lr=2e-5]\nTraining steps:  92%|█████████▏| 184/200 [1:25:46<07:36, 28.56s/it, grad_norm=0.00812, loss=0.114, lr=2e-5] \nTraining steps:  92%|█████████▎| 185/200 [1:26:13<06:59, 27.94s/it, grad_norm=0.00812, loss=0.114, lr=2e-5]\nTraining steps:  92%|█████████▎| 185/200 [1:26:13<06:59, 27.94s/it, grad_norm=0.0145, loss=0.119, lr=2e-5] \nTraining steps:  93%|█████████▎| 186/200 [1:26:38<06:20, 27.19s/it, grad_norm=0.0145, loss=0.119, lr=2e-5]\nTraining steps:  93%|█████████▎| 186/200 [1:26:38<06:20, 27.19s/it, grad_norm=0.00235, loss=0.206, lr=2e-5]\nTraining steps:  94%|█████████▎| 187/200 [1:27:10<06:11, 28.56s/it, grad_norm=0.00235, loss=0.206, lr=2e-5]\nTraining steps:  94%|█████████▎| 187/200 [1:27:10<06:11, 28.56s/it, grad_norm=0.00372, loss=0.0936, lr=2e-5]\nTraining steps:  94%|█████████▍| 188/200 [1:27:36<05:35, 27.97s/it, grad_norm=0.00372, loss=0.0936, lr=2e-5]\nTraining steps:  94%|█████████▍| 188/200 [1:27:36<05:35, 27.97s/it, grad_norm=0.0022, loss=0.136, lr=2e-5]  \nTraining steps:  94%|█████████▍| 189/200 [1:28:02<04:59, 27.20s/it, grad_norm=0.0022, loss=0.136, lr=2e-5]\nTraining steps:  94%|█████████▍| 189/200 [1:28:02<04:59, 27.20s/it, grad_norm=0.00224, loss=0.0354, lr=2e-5]\nTraining steps:  95%|█████████▌| 190/200 [1:28:34<04:46, 28.61s/it, grad_norm=0.00224, loss=0.0354, lr=2e-5]\nTraining steps:  95%|█████████▌| 190/200 [1:28:34<04:46, 28.61s/it, grad_norm=0.00368, loss=0.205, lr=2e-5] \nTraining steps:  96%|█████████▌| 191/200 [1:29:00<04:11, 27.97s/it, grad_norm=0.00368, loss=0.205, lr=2e-5]\nTraining steps:  96%|█████████▌| 191/200 [1:29:00<04:11, 27.97s/it, grad_norm=0.0022, loss=0.0938, lr=2e-5]\nTraining steps:  96%|█████████▌| 192/200 [1:29:26<03:37, 27.21s/it, grad_norm=0.0022, loss=0.0938, lr=2e-5]\nTraining steps:  96%|█████████▌| 192/200 [1:29:26<03:37, 27.21s/it, grad_norm=0.00187, loss=0.0347, lr=2e-5]\nTraining steps:  96%|█████████▋| 193/200 [1:29:58<03:20, 28.62s/it, grad_norm=0.00187, loss=0.0347, lr=2e-5]\nTraining steps:  96%|█████████▋| 193/200 [1:29:58<03:20, 28.62s/it, grad_norm=0.00241, loss=0.0955, lr=2e-5]\nTraining steps:  97%|█████████▋| 194/200 [1:30:24<02:47, 27.98s/it, grad_norm=0.00241, loss=0.0955, lr=2e-5]\nTraining steps:  97%|█████████▋| 194/200 [1:30:24<02:47, 27.98s/it, grad_norm=0.00195, loss=0.0951, lr=2e-5]\nTraining steps:  98%|█████████▊| 195/200 [1:30:49<02:16, 27.22s/it, grad_norm=0.00195, loss=0.0951, lr=2e-5]\nTraining steps:  98%|█████████▊| 195/200 [1:30:49<02:16, 27.22s/it, grad_norm=0.00134, loss=0.0393, lr=2e-5]\nTraining steps:  98%|█████████▊| 196/200 [1:31:21<01:54, 28.63s/it, grad_norm=0.00134, loss=0.0393, lr=2e-5]\nTraining steps:  98%|█████████▊| 196/200 [1:31:21<01:54, 28.63s/it, grad_norm=0.00359, loss=0.0926, lr=2e-5]\nTraining steps:  98%|█████████▊| 197/200 [1:31:48<01:23, 27.98s/it, grad_norm=0.00359, loss=0.0926, lr=2e-5]\nTraining steps:  98%|█████████▊| 197/200 [1:31:48<01:23, 27.98s/it, grad_norm=0.0022, loss=0.135, lr=2e-5]  \nTraining steps:  99%|█████████▉| 198/200 [1:32:13<00:54, 27.22s/it, grad_norm=0.0022, loss=0.135, lr=2e-5]\nTraining steps:  99%|█████████▉| 198/200 [1:32:13<00:54, 27.22s/it, grad_norm=0.024, loss=0.135, lr=2e-5] \nTraining steps: 100%|█████████▉| 199/200 [1:32:45<00:28, 28.64s/it, grad_norm=0.024, loss=0.135, lr=2e-5]\nTraining steps: 100%|█████████▉| 199/200 [1:32:45<00:28, 28.64s/it, grad_norm=0.00479, loss=0.0979, lr=2e-5]\nTraining steps: 100%|██████████| 200/200 [1:33:12<00:00, 27.99s/it, grad_norm=0.00479, loss=0.0979, lr=2e-5]\nTraining steps: 100%|██████████| 200/200 [1:33:12<00:00, 27.99s/it, grad_norm=0.00182, loss=0.0937, lr=2e-5]Model weights saved in output/pytorch_lora_weights.safetensors\n01/06/2025 20:39:45 - INFO - finetrainers - Starting validation\n01/06/2025 20:39:45 - WARNING - finetrainers - No validation samples found. Skipping validation.\n01/06/2025 20:39:46 - INFO - finetrainers - Memory after training end: {\n\"memory_allocated\": 25.392,\n\"memory_reserved\": 29.039,\n\"max_memory_allocated\": 55.037,\n\"max_memory_reserved\": 58.52\n}\nTraining steps: 100%|██████████| 200/200 [1:33:12<00:00, 27.96s/it, grad_norm=0.00182, loss=0.0937, lr=2e-5]\nThe token has not been saved to the git credentials helper. Pass `add_to_git_credential=True` in this function directly or `--add-to-git-credential` if using via `huggingface-cli` if you want to set the git credential as well.\nToken is valid (permission: write).\nThe token `Discord` has been saved to /root/.cache/huggingface/stored_tokens\nYour token has been saved to /root/.cache/huggingface/token\nLogin successful.\nThe current active token is: `Discord`\nConsider using `hf_transfer` for faster uploads. This solution comes with some limitations. See https://huggingface.co/docs/huggingface_hub/hf_transfer for more details.\nStart hashing 1 files.\nFinished hashing 1 files.\npytorch_lora_weights.safetensors:   0%|          | 0.00/327M [00:00<?, ?B/s]\npytorch_lora_weights.safetensors:   5%|▍         | 16.0M/327M [00:00<00:08, 34.9MB/s]\npytorch_lora_weights.safetensors:  10%|▉         | 32.0M/327M [00:00<00:05, 53.9MB/s]\npytorch_lora_weights.safetensors:  15%|█▍        | 48.0M/327M [00:02<00:17, 16.2MB/s]\npytorch_lora_weights.safetensors:  20%|█▉        | 64.0M/327M [00:02<00:11, 23.6MB/s]\npytorch_lora_weights.safetensors:  24%|██▍       | 80.0M/327M [00:02<00:07, 31.1MB/s]\npytorch_lora_weights.safetensors:  29%|██▉       | 96.0M/327M [00:03<00:05, 40.4MB/s]\npytorch_lora_weights.safetensors:  34%|███▍      | 112M/327M [00:03<00:04, 43.7MB/s] \npytorch_lora_weights.safetensors:  39%|███▉      | 128M/327M [00:03<00:03, 50.8MB/s]\npytorch_lora_weights.safetensors:  44%|████▍     | 144M/327M [00:03<00:03, 60.8MB/s]\npytorch_lora_weights.safetensors:  49%|████▉     | 160M/327M [00:03<00:02, 62.7MB/s]\npytorch_lora_weights.safetensors:  54%|█████▍    | 176M/327M [00:04<00:02, 63.8MB/s]\npytorch_lora_weights.safetensors:  59%|█████▊    | 192M/327M [00:04<00:02, 67.4MB/s]\npytorch_lora_weights.safetensors:  64%|██████▎   | 208M/327M [00:04<00:01, 71.9MB/s]\npytorch_lora_weights.safetensors:  68%|██████▊   | 224M/327M [00:05<00:02, 50.6MB/s]\npytorch_lora_weights.safetensors:  73%|███████▎  | 240M/327M [00:05<00:01, 52.6MB/s]\npytorch_lora_weights.safetensors:  78%|███████▊  | 256M/327M [00:05<00:01, 55.6MB/s]\npytorch_lora_weights.safetensors:  83%|████████▎ | 272M/327M [00:05<00:00, 60.7MB/s]\npytorch_lora_weights.safetensors:  88%|████████▊ | 288M/327M [00:06<00:00, 64.0MB/s]\npytorch_lora_weights.safetensors:  93%|█████████▎| 304M/327M [00:06<00:00, 65.8MB/s]\npytorch_lora_weights.safetensors:  98%|█████████▊| 320M/327M [00:06<00:00, 68.1MB/s]\npytorch_lora_weights.safetensors: 100%|██████████| 327M/327M [00:06<00:00, 49.4MB/s]\nhttps://huggingface.co/lucataco/hunyuan-lora-melty-test-3/tree/main/.\n./\n./pytorch_lora_weights.safetensors",
  "metrics": {
    "predict_time": 5636.337686541,
    "total_time": 5703.476216
  },
  "output": "https://replicate.delivery/xezq/uVj6v5243arXHhbgZq3tmDnfItI38VoDgoFyaxQOMsCOjPBKA/trained_model.tar",
  "started_at": "2025-01-06T19:06:00.291530Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-62ek6pwchg2xzwp4qpvj6534h4efntyo6aiqg32rxq7akx5wejrq",
    "get": "https://api.replicate.com/v1/predictions/6cdr553mm5rm80cm7m0vk6bhsr",
    "cancel": "https://api.replicate.com/v1/predictions/6cdr553mm5rm80cm7m0vk6bhsr/cancel"
  },
  "version": "153d95948aa1aa06b8dc413282337a6fba379af00eb3ea08af4938a2c03788d8"
}