Add files using upload-large-folder tool

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Dynamic_Spatial_Relationship.log

@@ -0,0 +1,21 @@
+[2026-06-08 19:14:26 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Dynamic_Spatial_Relationship --dimension Dynamic_Spatial_Relationship --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Dynamic_Spatial_Relationship', dimension=['Dynamic_Spatial_Relationship'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship/results_2026-06-08-19:14:29_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship/results_2026-06-08-19:14:29_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 19:14:35,488 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+
  0%|          | 0/69 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  1%|▏         | 1/69 [00:07<07:57,  7.01s/it]
  3%|▎         | 2/69 [00:13<07:09,  6.41s/it]
  4%|▍         | 3/69 [00:19<06:58,  6.35s/it]
  6%|▌         | 4/69 [00:25<06:44,  6.22s/it]
  7%|▋         | 5/69 [00:31<06:42,  6.29s/it]
  9%|▊         | 6/69 [00:37<06:30,  6.20s/it]
 10%|█         | 7/69 [00:43<06:19,  6.13s/it]
 12%|█▏        | 8/69 [00:50<06:20,  6.24s/it]
 13%|█▎        | 9/69 [00:56<06:08,  6.15s/it]
 14%|█▍        | 10/69 [01:02<06:06,  6.21s/it]
 16%|█▌        | 11/69 [01:08<05:56,  6.15s/it]
 17%|█▋        | 12/69 [01:14<05:50,  6.15s/it]
 19%|█▉        | 13/69 [01:20<05:42,  6.12s/it]
 20%|██        | 14/69 [01:26<05:34,  6.08s/it]
 22%|██▏       | 15/69 [01:32<05:31,  6.14s/it]
 23%|██▎       | 16/69 [01:39<05:25,  6.14s/it]
 25%|██▍       | 17/69 [01:45<05:17,  6.11s/it]
 26%|██▌       | 18/69 [01:51<05:13,  6.14s/it]
 28%|██▊       | 19/69 [01:57<05:03,  6.08s/it]
 29%|██▉       | 20/69 [02:03<04:58,  6.09s/it]
 30%|███       | 21/69 [02:09<04:51,  6.07s/it]
 32%|███▏      | 22/69 [02:15<04:46,  6.10s/it]
 33%|���██▎      | 23/69 [02:21<04:39,  6.07s/it]
 35%|███▍      | 24/69 [02:27<04:36,  6.14s/it]
 36%|███▌      | 25/69 [02:34<04:29,  6.13s/it]
 38%|███▊      | 26/69 [02:40<04:22,  6.10s/it]
 39%|███▉      | 27/69 [02:46<04:16,  6.10s/it]
 41%|████      | 28/69 [02:52<04:08,  6.07s/it]
 42%|████▏     | 29/69 [02:58<04:04,  6.11s/it]
 43%|████▎     | 30/69 [03:04<04:00,  6.17s/it]
 45%|████▍     | 31/69 [03:10<03:53,  6.15s/it]
 46%|████▋     | 32/69 [03:16<03:47,  6.14s/it]
 48%|████▊     | 33/69 [03:23<03:41,  6.14s/it]
 49%|████▉     | 34/69 [03:29<03:33,  6.11s/it]
 51%|█████     | 35/69 [03:35<03:29,  6.15s/it]
 52%|█████▏    | 36/69 [03:41<03:23,  6.16s/it]
 54%|█████▎    | 37/69 [03:47<03:19,  6.22s/it]
 55%|█████▌    | 38/69 [03:54<03:21,  6.50s/it]
 57%|█████▋    | 39/69 [04:01<03:16,  6.54s/it]
 58%|█████▊    | 40/69 [04:08<03:12,  6.63s/it]
 59%|█████▉    | 41/69 [04:14<03:02,  6.51s/it]
 61%|██████    | 42/69 [04:21<02:55,  6.51s/it]
 62%|██████▏   | 43/69 [04:28<02:56,  6.79s/it]
 64%|██████▍   | 44/69 [04:36<02:58,  7.15s/it]
 65%|██████▌   | 45/69 [04:44<02:58,  7.42s/it]
 67%|██████▋   | 46/69 [04:52<02:54,  7.60s/it]
 68%|██████▊   | 47/69 [05:00<02:48,  7.67s/it]
 70%|██████▉   | 48/69 [05:08<02:44,  7.83s/it]
 71%|███████   | 49/69 [05:16<02:37,  7.89s/it]
 72%|███████▏  | 50/69 [05:25<02:31,  7.99s/it]
 74%|███████▍  | 51/69 [05:33<02:25,  8.08s/it]
 75%|███████▌  | 52/69 [05:40<02:14,  7.92s/it]
 77%|███████▋  | 53/69 [05:48<02:06,  7.92s/it]
 78%|███████▊  | 54/69 [05:57<02:00,  8.03s/it]
 80%|███████▉  | 55/69 [06:05<01:53,  8.10s/it]
 81%|████████  | 56/69 [06:13<01:46,  8.16s/it]
 83%|████████▎ | 57/69 [06:21<01:37,  8.09s/it]
 84%|████████▍ | 58/69 [06:29<01:28,  8.04s/it]
 86%|████████▌ | 59/69 [06:37<01:20,  8.09s/it]
 87%|████████▋ | 60/69 [06:45<01:12,  8.01s/it]
 88%|████████▊ | 61/69 [06:53<01:03,  7.94s/it]
 90%|████████▉ | 62/69 [07:01<00:56,  8.07s/it]
 91%|█████████▏| 63/69 [07:09<00:48,  8.01s/it]
 93%|█████████▎| 64/69 [07:18<00:40,  8.17s/it]
 94%|█████████▍| 65/69 [07:26<00:32,  8.24s/it]
 96%|█████████▌| 66/69 [07:35<00:25,  8.35s/it]
 97%|█████████▋| 67/69 [07:43<00:16,  8.33s/it]
 99%|█████████▊| 68/69 [07:51<00:08,  8.32s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship/results_2026-06-08-19:14:29_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Human_Interaction.log

@@ -0,0 +1,23 @@
+[2026-06-08 17:28:50 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Interaction --dimension Human_Interaction --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Interaction', dimension=['Human_Interaction'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+File exists: /root/ckpts/vbench2/Qwen/Qwen2.5-7B-Instruct
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction/results_2026-06-08-17:28:53_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction/results_2026-06-08-17:28:53_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+
+Model Class: LlavaQwenForCausalLM
+2026-06-08 17:29:04,656 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+
  0%|          | 0/100 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  1%|          | 1/100 [00:42<1:09:20, 42.02s/it]
  2%|▏         | 2/100 [01:23<1:08:34, 41.99s/it]
  3%|▎         | 3/100 [02:00<1:04:07, 39.66s/it]
  4%|▍         | 4/100 [02:38<1:02:00, 38.75s/it]
  5%|▌         | 5/100 [03:22<1:04:37, 40.81s/it]
  6%|▌         | 6/100 [04:01<1:03:04, 40.26s/it]
  7%|▋         | 7/100 [04:47<1:04:51, 41.85s/it]
  8%|▊         | 8/100 [05:51<1:15:07, 48.99s/it]
  9%|▉         | 9/100 [06:52<1:19:56, 52.71s/it]
 10%|█         | 10/100 [07:47<1:20:21, 53.57s/it]
 11%|█         | 11/100 [08:39<1:18:48, 53.13s/it]
 12%|█▏        | 12/100 [09:40<1:21:20, 55.46s/it]
 13%|█▎        | 13/100 [10:45<1:24:46, 58.46s/it]
 14%|█▍        | 14/100 [11:45<1:24:05, 58.67s/it]
 15%|█▌        | 15/100 [12:38<1:20:51, 57.08s/it]
 16%|█▌        | 16/100 [13:30<1:17:48, 55.58s/it]
 17%|█▋        | 17/100 [14:09<1:10:07, 50.70s/it]
 18%|█▊        | 18/100 [14:48<1:04:13, 47.00s/it]
 19%|█▉        | 19/100 [15:31<1:01:46, 45.76s/it]
 20%|██        | 20/100 [16:08<57:41, 43.27s/it]  
 21%|██        | 21/100 [16:48<55:29, 42.15s/it]
 22%|██▏       | 22/100 [17:26<53:25, 41.10s/it]
 23%|██▎       | 23/100 [18:11<53:56, 42.04s/it]
 24%|██▍       | 24/100 [19:04<57:37, 45.49s/it]
 25%|██▌       | 25/100 [19:59<1:00:12, 48.16s/it]
 26%|██▌       | 26/100 [20:59<1:03:54, 51.82s/it]
 27%|██▋       | 27/100 [22:02<1:07:19, 55.34s/it]
 28%|██▊       | 28/100 [22:56<1:05:51, 54.88s/it]
 29%|██▉       | 29/100 [23:59<1:07:34, 57.11s/it]
 30%|███       | 30/100 [24:52<1:05:30, 56.16s/it]
 31%|███       | 31/100 [25:47<1:03:57, 55.62s/it]
 32%|███▏      | 32/100 [26:36<1:00:52, 53.71s/it]
 33%|███▎      | 33/100 [27:20<56:42, 50.78s/it]  
 34%|███▍      | 34/100 [28:01<52:30, 47.74s/it]
 35%|███▌      | 35/100 [28:42<49:40, 45.86s/it]
 36%|███▌      | 36/100 [29:23<47:26, 44.48s/it]
 37%|███▋      | 37/100 [30:15<48:50, 46.52s/it]
 38%|███▊      | 38/100 [30:59<47:18, 45.79s/it]
 39%|███▉      | 39/100 [31:36<44:03, 43.33s/it]
 40%|████      | 40/100 [32:29<45:58, 45.97s/it]
 41%|████      | 41/100 [33:26<48:36, 49.43s/it]
 42%|████▏     | 42/100 [34:20<49:06, 50.80s/it]
 43%|████▎     | 43/100 [35:30<53:51, 56.70s/it]
 44%|████▍     | 44/100 [36:31<53:57, 57.82s/it]
 45%|████▌     | 45/100 [37:32<53:48, 58.70s/it]
 46%|████▌     | 46/100 [38:25<51:28, 57.19s/it]
 47%|████▋     | 47/100 [39:26<51:19, 58.10s/it]
 48%|████▊     | 48/100 [40:12<47:23, 54.69s/it]
 49%|████▉     | 49/100 [40:53<42:49, 50.39s/it]
 50%|█████     | 50/100 [41:29<38:23, 46.07s/it]
 51%|█████     | 51/100 [42:14<37:27, 45.86s/it]
 52%|█████▏    | 52/100 [42:52<34:46, 43.46s/it]
 53%|█████▎    | 53/100 [43:28<32:23, 41.36s/it]
 54%|█████▍    | 54/100 [44:04<30:26, 39.71s/it]
 55%|█████▌    | 55/100 [44:44<29:55, 39.89s/it]
 56%|█████▌    | 56/100 [45:36<31:42, 43.23s/it]
 57%|█████▋    | 57/100 [46:43<36:07, 50.41s/it]
 58%|█████▊    | 58/100 [47:31<34:52, 49.83s/it]
 59%|█████▉    | 59/100 [48:45<38:55, 56.97s/it]
 60%|██████    | 60/100 [49:42<37:59, 56.99s/it]
 61%|██████    | 61/100 [50:41<37:26, 57.59s/it]
 62%|██████▏   | 62/100 [51:37<36:06, 57.02s/it]
 63%|██████▎   | 63/100 [52:29<34:20, 55.68s/it]
 64%|██████▍   | 64/100 [53:10<30:42, 51.17s/it]
 65%|██████▌   | 65/100 [53:51<28:04, 48.11s/it]
 66%|██████▌   | 66/100 [54:33<26:16, 46.36s/it]
 67%|██████▋   | 67/100 [55:17<25:09, 45.73s/it]
 68%|██████▊   | 68/100 [55:58<23:32, 44.13s/it]
 69%|██████▉   | 69/100 [56:37<22:06, 42.78s/it]
 70%|███████   | 70/100 [57:13<20:19, 40.66s/it]
 71%|███████   | 71/100 [58:11<22:09, 45.84s/it]
 72%|███████▏  | 72/100 [59:07<22:52, 49.02s/it]
 73%|███████▎  | 73/100 [1:00:06<23:25, 52.06s/it]
 74%|███████▍  | 74/100 [1:01:04<23:17, 53.74s/it]
 75%|███████▌  | 75/100 [1:02:20<25:09, 60.39s/it]
 76%|███████▌  | 76/100 [1:03:15<23:27, 58.66s/it]
 77%|███████▋  | 77/100 [1:04:22<23:26, 61.15s/it]
 78%|███████▊  | 78/100 [1:05:09<20:51, 56.90s/it]
 79%|███████▉  | 79/100 [1:05:58<19:09, 54.73s/it]
 80%|████████  | 80/100 [1:06:37<16:38, 49.92s/it]
 81%|████████  | 81/100 [1:07:25<15:38, 49.41s/it]
 82%|████████▏ | 82/100 [1:08:04<13:52, 46.24s/it]
 83%|████████▎ | 83/100 [1:08:47<12:49, 45.25s/it]
 84%|████████▍ | 84/100 [1:09:28<11:46, 44.13s/it]
 85%|████████▌ | 85/100 [1:10:08<10:39, 42.63s/it]
 86%|████████▌ | 86/100 [1:10:58<10:28, 44.88s/it]
 87%|████████▋ | 87/100 [1:12:06<11:14, 51.87s/it]
 88%|████████▊ | 88/100 [1:13:05<10:47, 53.96s/it]
 89%|████████▉ | 89/100 [1:14:14<10:43, 58.52s/it]
 90%|█████████ | 90/100 [1:15:30<10:36, 63.65s/it]
 91%|█████████ | 91/100 [1:16:28<09:18, 62.09s/it]
 92%|█████████▏| 92/100 [1:17:30<08:15, 62.00s/it]
 93%|█████████▎| 93/100 [1:18:28<07:05, 60.81s/it]
 94%|█████████▍| 94/100 [1:19:07<05:26, 54.37s/it]
 95%|█████████▌| 95/100 [1:19:47<04:09, 49.99s/it]
 96%|█████████▌| 96/100 [1:20:26<03:06, 46.65s/it]
 97%|█████████▋| 97/100 [1:21:10<02:17, 45.80s/it]
 98%|█████████▊| 98/100 [1:21:48<01:27, 43.55s/it]
 99%|█████████▉| 99/100 [1:22:34<00:44, 44.38s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction/results_2026-06-08-17:28:53_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_2.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_2 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_2.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_2', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_2.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_2/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_2/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/7 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 14%|█▍        | 1/7 [02:05<12:35, 125.86s/it]
 29%|██▊       | 2/7 [04:08<10:19, 123.98s/it]
 43%|████▎     | 3/7 [06:10<08:12, 123.16s/it]
 57%|█████▋    | 4/7 [08:10<06:05, 121.74s/it]
 71%|███████▏  | 5/7 [10:09<04:01, 120.98s/it]
 86%|████████▌ | 6/7 [12:09<02:00, 120.58s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_2/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_5.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_5 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_5.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_5', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_5.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_5/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_5/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/7 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 14%|█▍        | 1/7 [02:05<12:30, 125.11s/it]
 29%|██▊       | 2/7 [04:08<10:21, 124.26s/it]
 43%|████▎     | 3/7 [06:10<08:12, 123.18s/it]
 57%|█████▋    | 4/7 [08:10<06:06, 122.03s/it]
 71%|███████▏  | 5/7 [10:10<04:01, 120.96s/it]
 86%|████████▌ | 6/7 [12:09<02:00, 120.38s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_5/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Mechanics.log

@@ -0,0 +1,21 @@
+[2026-06-08 19:23:27 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Mechanics --dimension Mechanics --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Mechanics', dimension=['Mechanics'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics/results_2026-06-08-19:23:32_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics/results_2026-06-08-19:23:32_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 19:23:44,088 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+
  0%|          | 0/57 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  2%|▏         | 1/57 [00:19<17:46, 19.04s/it]
  4%|▎         | 2/57 [00:37<17:01, 18.58s/it]
  5%|▌         | 3/57 [00:57<17:19, 19.25s/it]
  7%|▋         | 4/57 [01:15<16:45, 18.96s/it]
  9%|▉         | 5/57 [01:34<16:17, 18.80s/it]
 11%|█         | 6/57 [01:52<15:44, 18.52s/it]
 12%|█▏        | 7/57 [02:10<15:21, 18.44s/it]
 14%|█▍        | 8/57 [02:28<14:49, 18.15s/it]
 16%|█▌        | 9/57 [02:37<12:26, 15.54s/it]
 18%|█▊        | 10/57 [02:48<10:53, 13.90s/it]
 19%|█▉        | 11/57 [02:58<09:54, 12.92s/it]
 21%|██        | 12/57 [03:09<09:08, 12.19s/it]
 23%|██▎       | 13/57 [03:25<09:50, 13.42s/it]
 25%|██▍       | 14/57 [03:35<08:47, 12.27s/it]
 26%|██▋       | 15/57 [03:44<07:59, 11.41s/it]
 28%|██▊       | 16/57 [04:00<08:43, 12.78s/it]
 30%|██▉       | 17/57 [04:16<09:09, 13.73s/it]
 32%|███▏      | 18/57 [04:32<09:25, 14.50s/it]
 33%|███▎      | 19/57 [04:48<09:26, 14.92s/it]
 35%|███▌      | 20/57 [05:04<09:24, 15.24s/it]
 37%|███▋      | 21/57 [05:20<09:17, 15.50s/it]
 39%|███▊      | 22/57 [05:30<07:58, 13.66s/it]
 40%|████      | 23/57 [05:39<07:03, 12.46s/it]
 42%|████▏     | 24/57 [05:49<06:20, 11.53s/it]
 44%|████▍     | 25/57 [05:59<05:52, 11.00s/it]
 46%|████▌     | 26/57 [06:08<05:25, 10.51s/it]
 47%|████▋     | 27/57 [06:17<05:05, 10.17s/it]
 49%|████▉     | 28/57 [06:27<04:47,  9.90s/it]
 51%|█████     | 29/57 [06:36<04:32,  9.75s/it]
 53%|█████▎    | 30/57 [06:52<05:12, 11.57s/it]
 54%|█████▍    | 31/57 [07:08<05:35, 12.89s/it]
 56%|█████▌    | 32/57 [07:24<05:45, 13.82s/it]
 58%|█████▊    | 33/57 [07:40<05:49, 14.58s/it]
 60%|█████▉    | 34/57 [07:57<05:50, 15.25s/it]
 61%|██████▏   | 35/57 [08:14<05:48, 15.86s/it]
 63%|██████▎   | 36/57 [08:31<05:40, 16.22s/it]
 65%|██████▍   | 37/57 [08:49<05:34, 16.71s/it]
 67%|██████▋   | 38/57 [09:07<05:23, 17.02s/it]
 68%|██████▊   | 39/57 [09:25<05:11, 17.28s/it]
 70%|███████   | 40/57 [09:42<04:55, 17.40s/it]
 72%|███████▏  | 41/57 [10:00<04:38, 17.38s/it]
 74%|███████▎  | 42/57 [10:18<04:23, 17.54s/it]
 75%|███████▌  | 43/57 [10:36<04:08, 17.78s/it]
 77%|███████▋  | 44/57 [10:54<03:50, 17.75s/it]
 79%|███████▉  | 45/57 [11:12<03:33, 17.79s/it]
 81%|████████  | 46/57 [11:30<03:17, 17.92s/it]
 82%|████████▏ | 47/57 [11:48<03:01, 18.12s/it]
 84%|████████▍ | 48/57 [12:06<02:41, 17.97s/it]
 86%|████████▌ | 49/57 [12:24<02:24, 18.07s/it]
 88%|████████▊ | 50/57 [12:42<02:06, 18.11s/it]
 89%|████████▉ | 51/57 [13:01<01:48, 18.13s/it]
 91%|█████████ | 52/57 [13:18<01:30, 18.04s/it]
 93%|█████████▎| 53/57 [13:36<01:11, 17.93s/it]
 95%|█████████▍| 54/57 [13:54<00:53, 17.99s/it]
 96%|█████████▋| 55/57 [14:12<00:35, 17.95s/it]
 98%|█████████▊| 56/57 [14:30<00:17, 17.83s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics/results_2026-06-08-19:23:32_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_score_details.csv

@@ -0,0 +1,25 @@
+model_name,dimension,score,score_percent,result_file
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Camera Motion,0.5740740740740741,57.407407407407405,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion/results_2026-06-08-18:42:29_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Complex Landscape,0.19999999999999987,19.999999999999986,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape/results_2026-06-08-20:58:55_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Complex Plot,0.125185185185185,12.518518518518501,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot/results_2026-06-08-19:23:02_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Composition,0.458385744234801,45.8385744234801,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Composition/results_2026-06-08-20:51:55_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Diversity,0.3863112598657608,38.63112598657608,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity/results_2026-06-08-17:28:53_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Dynamic Attribute,0.24175824175824176,24.175824175824175,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute/results_2026-06-08-18:53:30_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Dynamic Spatial Relationship,0.3140096618357488,31.40096618357488,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship/results_2026-06-08-19:14:29_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Human Anatomy,0.932573792689079,93.2573792689079,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy/results_2026-06-08-17:28:53_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Human Clothes,0.9620853080568721,96.2085308056872,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes/results_2026-06-08-20:23:53_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Human Identity,0.850909343083323,85.0909343083323,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity/results_2026-06-08-20:25:23_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Human Interaction,0.49666666666666665,49.666666666666664,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction/results_2026-06-08-17:28:53_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Instance Preservation,0.935672514619883,93.56725146198829,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation/results_2026-06-08-21:49:42_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Material,0.46534653465346537,46.53465346534654,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material/results_2026-06-08-19:56:51_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Mechanics,0.5210084033613446,52.10084033613446,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics/results_2026-06-08-19:23:32_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Motion Order Understanding,0.16835016835016836,16.835016835016837,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding/results_2026-06-08-18:01:56_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Motion Rationality,0.3620689655172414,36.206896551724135,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Rationality/results_2026-06-08-19:56:51_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Multi-View Consistency,0.3463625062863282,34.63625062863282,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency/results_2026-06-08-17:28:53_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Thermotics,0.4413793103448276,44.13793103448276,/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics/results_2026-06-08-20:10:22_eval_results.json
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Creativity,0.4223485020502809,42.23485020502809,
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Commonsense,0.6488707400685622,64.88707400685621,
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Controllability,0.30286342826715495,30.286342826715497,
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Human Fidelity,0.9151894812764247,91.51894812764247,
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Physics,0.44352418866149146,44.352418866149144,
+yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100,Overall,0.5465592680647828,54.655926806478284,

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_score_summary.json

@@ -0,0 +1,82 @@
+{
+  "model_name": "yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting_step_100",
+  "generated_at_utc": "2026-06-08T22:11:45.382809+00:00",
+  "results_root": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results",
+  "score_formula": "mean of five VBench2 groups; each group is the arithmetic mean of its listed dimensions",
+  "dimension_scores": {
+    "Camera Motion": 0.5740740740740741,
+    "Complex Landscape": 0.19999999999999987,
+    "Complex Plot": 0.125185185185185,
+    "Composition": 0.458385744234801,
+    "Diversity": 0.3863112598657608,
+    "Dynamic Attribute": 0.24175824175824176,
+    "Dynamic Spatial Relationship": 0.3140096618357488,
+    "Human Anatomy": 0.932573792689079,
+    "Human Clothes": 0.9620853080568721,
+    "Human Identity": 0.850909343083323,
+    "Human Interaction": 0.49666666666666665,
+    "Instance Preservation": 0.935672514619883,
+    "Material": 0.46534653465346537,
+    "Mechanics": 0.5210084033613446,
+    "Motion Order Understanding": 0.16835016835016836,
+    "Motion Rationality": 0.3620689655172414,
+    "Multi-View Consistency": 0.3463625062863282,
+    "Thermotics": 0.4413793103448276
+  },
+  "group_scores": {
+    "Creativity": 0.4223485020502809,
+    "Commonsense": 0.6488707400685622,
+    "Controllability": 0.30286342826715495,
+    "Human Fidelity": 0.9151894812764247,
+    "Physics": 0.44352418866149146
+  },
+  "overall": 0.5465592680647828,
+  "dimension_scores_percent": {
+    "Camera Motion": 57.407407407407405,
+    "Complex Landscape": 19.999999999999986,
+    "Complex Plot": 12.518518518518501,
+    "Composition": 45.8385744234801,
+    "Diversity": 38.63112598657608,
+    "Dynamic Attribute": 24.175824175824175,
+    "Dynamic Spatial Relationship": 31.40096618357488,
+    "Human Anatomy": 93.2573792689079,
+    "Human Clothes": 96.2085308056872,
+    "Human Identity": 85.0909343083323,
+    "Human Interaction": 49.666666666666664,
+    "Instance Preservation": 93.56725146198829,
+    "Material": 46.53465346534654,
+    "Mechanics": 52.10084033613446,
+    "Motion Order Understanding": 16.835016835016837,
+    "Motion Rationality": 36.206896551724135,
+    "Multi-View Consistency": 34.63625062863282,
+    "Thermotics": 44.13793103448276
+  },
+  "group_scores_percent": {
+    "Creativity": 42.23485020502809,
+    "Commonsense": 64.88707400685621,
+    "Controllability": 30.286342826715497,
+    "Human Fidelity": 91.51894812764247,
+    "Physics": 44.352418866149144
+  },
+  "overall_percent": 54.655926806478284,
+  "result_files": {
+    "Camera Motion": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion/results_2026-06-08-18:42:29_eval_results.json",
+    "Complex Landscape": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape/results_2026-06-08-20:58:55_eval_results.json",
+    "Complex Plot": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot/results_2026-06-08-19:23:02_eval_results.json",
+    "Composition": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Composition/results_2026-06-08-20:51:55_eval_results.json",
+    "Diversity": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity/results_2026-06-08-17:28:53_eval_results.json",
+    "Dynamic Attribute": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute/results_2026-06-08-18:53:30_eval_results.json",
+    "Dynamic Spatial Relationship": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship/results_2026-06-08-19:14:29_eval_results.json",
+    "Human Anatomy": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy/results_2026-06-08-17:28:53_eval_results.json",
+    "Human Clothes": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes/results_2026-06-08-20:23:53_eval_results.json",
+    "Human Identity": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity/results_2026-06-08-20:25:23_eval_results.json",
+    "Human Interaction": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction/results_2026-06-08-17:28:53_eval_results.json",
+    "Instance Preservation": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation/results_2026-06-08-21:49:42_eval_results.json",
+    "Material": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material/results_2026-06-08-19:56:51_eval_results.json",
+    "Mechanics": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics/results_2026-06-08-19:23:32_eval_results.json",
+    "Motion Order Understanding": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding/results_2026-06-08-18:01:56_eval_results.json",
+    "Motion Rationality": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Rationality/results_2026-06-08-19:56:51_eval_results.json",
+    "Multi-View Consistency": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency/results_2026-06-08-17:28:53_eval_results.json",
+    "Thermotics": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics/results_2026-06-08-20:10:22_eval_results.json"
+  }
+}

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260530_214851-dgy7zq92/files/requirements.txt

@@ -0,0 +1,283 @@
+awscli==1.45.3
+boto3==1.43.3
+botocore==1.43.3
+certifi==2026.4.22
+charset-normalizer==3.4.7
+colorama==0.4.6
+docutils==0.19
+idna==3.13
+jmespath==1.1.0
+pip==26.1.1
+pyasn1==0.6.3
+python-dateutil==2.9.0.post0
+PyYAML==6.0.3
+requests==2.33.1
+rsa==4.7.2
+s3transfer==0.17.0
+six==1.17.0
+urllib3==2.6.3
+py-cpuinfo==9.0.0
+hjson==3.1.0
+easydict==1.13
+braceexpand==0.1.7
+antlr4-python3-runtime==4.9.3
+websocket-client==1.9.0
+webdataset==1.0.2
+triton==3.6.0
+torchcodec==0.13.0+cu130
+tifffile==2026.5.15
+tenacity==9.1.4
+smmap==5.0.3
+sentry-sdk==2.61.0
+sentencepiece==0.2.1
+safetensors==0.7.0
+regex==2026.5.9
+pyarrow==24.0.0
+pyaml==26.2.1
+propcache==0.5.2
+opencv-python-headless==4.13.0.92
+omegaconf==2.3.0
+lightning-utilities==0.15.3
+nvidia-nvtx==13.0.85
+nvidia-nvshmem-cu13==3.4.5
+lazy-loader==0.5
+nvidia-nvjitlink==13.0.88
+nvidia-nccl-cu13==2.28.9
+imageio-ffmpeg==0.6.0
+nvidia-cufile==1.15.1.6
+nvidia-cuda-runtime==13.0.96
+cuda-pathfinder==1.5.5
+nvidia-cuda-nvrtc==13.0.88
+nvidia-cuda-cupti==13.0.85
+nvidia-cublas==13.1.0.3
+multidict==6.7.1
+msgpack==1.1.2
+Mako==1.3.12
+loguru==0.7.3
+lmdb==2.2.0
+ImageIO==2.37.3
+greenlet==3.5.1
+future==1.0.0
+ftfy==6.3.1
+frozenlist==1.8.0
+frozendict==2.4.7
+docstring_parser==0.18.0
+decord==0.6.0
+av==17.0.1
+attrs==26.1.0
+aiohappyeyeballs==2.6.2
+yarl==1.24.2
+typed-argument-parser==1.12.0
+SQLAlchemy==2.0.50
+scikit-image==0.26.0
+qwen-vl-utils==0.0.14
+nvidia-cusparse==12.6.3.3
+xxhash==3.7.0
+nvidia-cufft==12.0.0.61
+flash_attn==2.8.3+cu130torch2.10
+nvidia-cudnn-cu13==9.15.1.9
+huggingface_hub==0.36.2
+gitdb==4.0.12
+ffmpeg-python==0.2.0
+einx==0.4.3
+cupy-cuda12x==14.1.0
+cuda-bindings==13.0.3
+cattrs==26.1.0
+aiosignal==1.4.0
+tokenizers==0.21.4
+typeguard==4.5.2
+nvidia-cusolver==12.0.4.66
+mediapy==1.2.6
+ipdb==0.13.13
+GitPython==3.1.50
+diffusers==0.38.0
+coremltools==9.0
+alembic==1.18.4
+aiohttp==3.13.5
+wandb==0.27.0
+sniffio==1.3.1
+torch==2.10.0+cu130
+dataset==2.0.0
+dashscope==1.25.19
+torchvision==0.25.0+cu130
+torchmetrics==1.9.0
+torchdata==0.8.0
+termcolor==3.3.0
+torchaudio==2.10.0+cu130
+deepspeed==0.19.0
+timm==1.0.27
+pytorch-lightning==2.6.5
+peft==0.19.1
+lpips==0.1.4
+lightning==2.6.5
+shortuuid==1.0.13
+setuptools==82.0.0
+pillow==12.1.1
+numpy==2.4.2
+jiter==0.15.0
+fsspec==2026.2.0
+distro==1.9.0
+dill==0.4.1
+tyro==1.0.13
+tiktoken==0.13.0
+multiprocess==0.70.19
+modelscope==1.36.0
+fire==0.7.0
+openai==2.8.0
+datasets==4.8.4
+trl==0.8.6
+angelslim==0.2.2
+transformers==4.53.2
+black==26.3.1
+pathspec==1.1.1
+pytokens==0.4.1
+importlib_metadata==9.0.0
+ml_dtypes==0.5.4
+onnx==1.21.0
+onnx-ir==0.2.1
+onnxscript==0.7.0
+transformer_engine==2.9.0+70f53666
+zipp==3.23.1
+nvidia-mathdx==25.6.0
+einops==0.8.2
+absl-py==2.4.0
+accelerate==1.13.0
+annotated-doc==0.0.4
+annotated-types==0.7.0
+anyio==4.13.0
+apsw==3.53.1.0
+apswutils==0.1.2
+beartype==0.22.9
+beautifulsoup4==4.14.3
+blis==1.2.1
+catalogue==2.0.10
+cloudpathlib==0.24.0
+cloudpickle==3.1.2
+confection==0.1.5
+contourpy==1.3.3
+cycler==0.12.1
+cymem==2.0.13
+fastai==2.8.7
+fastcore==1.12.47
+fastdownload==0.0.7
+fastlite==0.2.4
+fastprogress==1.1.6
+fasttransform==0.0.2
+fonttools==4.62.1
+grpcio==1.80.0
+h11==0.16.0
+hf-xet==1.5.0
+httpcore==1.0.9
+httptools==0.7.1
+httpx==0.28.1
+itsdangerous==2.2.0
+joblib==1.5.3
+kiwisolver==1.5.0
+langcodes==3.5.1
+Markdown==3.10.2
+markdown-it-py==4.2.0
+matplotlib==3.10.9
+mdurl==0.1.2
+mpmath==1.3.0
+murmurhash==1.0.15
+mypy_extensions==1.1.0
+networkx==3.6.1
+nvidia-curand==10.4.0.35
+nvidia-cusparselt-cu13==0.8.0
+oauthlib==3.3.1
+pandas==3.0.2
+plum-dispatch==2.9.0
+preshed==3.0.13
+protobuf==7.34.1
+pydantic==2.13.4
+pydantic_core==2.46.4
+pyparsing==3.3.2
+pyre-extensions==0.0.32
+python-dotenv==1.2.2
+python-fasthtml==0.14.1
+python-multipart==0.0.28
+rich==15.0.0
+s3torchconnector==1.5.0
+s3torchconnectorclient==1.5.0
+scikit-learn==1.8.0
+shellingham==1.5.4
+smart_open==7.6.1
+soupsieve==2.8.3
+spacy==3.8.7
+spacy-legacy==3.0.12
+spacy-loggers==1.0.5
+srsly==2.5.3
+starlette==1.0.0
+sympy==1.14.0
+tabulate==0.10.0
+tensorboard==2.20.0
+tensorboard-data-server==0.7.2
+thinc==8.3.4
+threadpoolctl==3.6.0
+torchtnt==0.2.4
+typer==0.25.1
+typer-slim==0.24.0
+typing-inspect==0.9.0
+typing-inspection==0.4.2
+uvicorn==0.46.0
+uvloop==0.22.1
+wasabi==1.1.3
+watchfiles==1.1.1
+weasel==0.4.3
+websockets==16.0
+Werkzeug==3.1.8
+wrapt==2.1.2
+asttokens==3.0.1
+cffi==2.0.0
+click==8.3.3
+comm==0.2.3
+cryptography==48.0.0
+Cython==3.2.4
+debugpy==1.8.20
+decorator==5.2.1
+executing==2.2.1
+filelock==3.29.0
+h5py==3.16.0
+intel-cmplr-lib-ur==2026.0.0
+intel-openmp==2026.0.0
+ipykernel==7.2.0
+ipython==9.13.0
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+Jinja2==3.1.6
+jupyter_client==8.8.0
+jupyter_core==5.9.1
+MarkupSafe==3.0.3
+matplotlib-inline==0.2.2
+mkl==2026.0.0
+mkl-include==2026.0.0
+mpi4py==4.1.1
+nest-asyncio==1.6.0
+ninja==1.13.0
+onemkl-license==2026.0.0
+opencv-python==4.11.0.86
+packaging==26.2
+parso==0.8.7
+pexpect==4.9.0
+platformdirs==4.9.6
+prompt_toolkit==3.0.52
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+pybind11==3.0.4
+pycparser==3.0
+Pygments==2.20.0
+pyOpenSSL==26.2.0
+pyzmq==27.1.0
+scipy==1.17.1
+stack-data==0.6.3
+tbb==2023.0.0
+tcmlib==1.5.0
+tornado==6.5.5
+tqdm==4.67.3
+traitlets==5.15.0
+typing_extensions==4.15.0
+tzdata==2026.2
+umf==1.1.0
+wheel==0.47.0
+wcwidth==0.7.0

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260530_214851-dgy7zq92/files/wandb-metadata.json

@@ -0,0 +1,167 @@
+{
+  "os": "Linux-6.1.163-186.299.amzn2023.x86_64-x86_64-with-glibc2.35",
+  "python": "CPython 3.12.10",
+  "startedAt": "2026-05-30T21:48:51.336923Z",
+  "args": [
+    "--pretrained_model_root",
+    "./ckpts",
+    "--train_video_csv",
+    "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/assets/demo_train_decomposed_binary5.csv",
+    "--reward_model",
+    "video_score2",
+    "--reward_weights",
+    "{\"VQ\":1.0,\"MQ\":1.0,\"TA\":1.0}",
+    "--reward_gradient_credit",
+    "True",
+    "--reward_credit_num_frames",
+    "8",
+    "--reward_credit_max_questions",
+    "1",
+    "--reward_credit_metric",
+    "weighted",
+    "--token_credit_loss_mode",
+    "video_adv",
+    "--token_credit_ratio_mode",
+    "uniform_weighted_logprob",
+    "--auto_resume_from_timer",
+    "True",
+    "--learning_rate",
+    "1e-5",
+    "--batch_size",
+    "2",
+    "--num_generations",
+    "4",
+    "--max_steps",
+    "10000",
+    "--output_dir",
+    "./outputs",
+    "--enable_wandb",
+    "True",
+    "--project_name",
+    "SAGE-GRPO",
+    "--run_name",
+    "yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting",
+    "--enable_fsdp",
+    "--enable_gradient_checkpointing",
+    "--sp_size",
+    "2",
+    "--reward_checkpoint_mode",
+    "v3",
+    "--validation_interval",
+    "0",
+    "--validate_at_step0",
+    "False",
+    "--validate_video_length",
+    "121",
+    "--validation_timestep_shift",
+    "5.0",
+    "--use_grad_balancing",
+    "True",
+    "--enable_timestep_permutation",
+    "True",
+    "--sde_type",
+    "sage_grpo",
+    "--kl_weight",
+    "1e-5",
+    "--kl_coef",
+    "1e-7",
+    "--use_moving_KL",
+    "True",
+    "--update_ref_model_step",
+    "10",
+    "--use_dual_kl",
+    "True",
+    "--dual_kl_moving_weight",
+    "1.0",
+    "--dual_kl_step_weight",
+    "0.1",
+    "--reference_mode_offload",
+    "True",
+    "--checkpoints_directory",
+    "s3://snap-genvid-us-east-2/ywang29/checkpoints_large_model/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting"
+  ],
+  "program": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/post_train.py",
+  "codePath": "post_train.py",
+  "codePathLocal": "post_train.py",
+  "git": {
+    "remote": "https://github.com/Tencent-Hunyuan/SAGE-GRPO.git",
+    "commit": "779776e42732e0262b3f2844dd206d9d8f043b2f"
+  },
+  "root": "./outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb",
+  "host": "ywang29-hunyuan-sage-vs2-tcva-abl-nw-worker-0",
+  "username": "yi-fan-wang1216",
+  "executable": "/usr/local/bin/python",
+  "cpu_count": 48,
+  "cpu_count_logical": 96,
+  "gpu": "NVIDIA A100-SXM4-80GB",
+  "gpu_count": 8,
+  "disk": {
+    "/": {
+      "total": "7516777873408",
+      "used": "125919944704"
+    }
+  },
+  "memory": {
+    "total": "1204539265024"
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-5d116268-210b-57db-2b03-f1aabeab6fd2"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-b23d9c11-79de-3572-4a00-db7bd0e9b70d"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-4f20ada8-fcc0-b260-b094-0761ca5e508f"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-cd1a414c-3b66-bffa-5d71-4d05864102d8"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-5aa5a39d-c2d7-a96e-7f70-ff385c294b9d"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-6e5af7bd-1fda-e18b-8099-9a420b5fa901"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-2d5becf4-3167-7149-b346-42a95d41af41"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-b39fb2e0-d263-ba3f-7517-808e2ec43b86"
+    }
+  ],
+  "cudaVersion": "13.0",
+  "writerId": "ov4fgpb453fgr52zxb6m8bslzqexuh5l"
+}

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260530_214851-dgy7zq92/logs/debug.log

@@ -0,0 +1,20 @@
+2026-05-30 21:48:51,357 INFO    MainThread:1060908 [wandb_setup.py:_flush():81] Current SDK version is 0.27.0
+2026-05-30 21:48:51,358 INFO    MainThread:1060908 [wandb_setup.py:_flush():81] Configure stats pid to 1060908
+2026-05-30 21:48:51,358 INFO    MainThread:1060908 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-05-30 21:48:51,358 INFO    MainThread:1060908 [wandb_init.py:setup_run_log_directory():723] Logging user logs to outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260530_214851-dgy7zq92/logs/debug.log
+2026-05-30 21:48:51,358 INFO    MainThread:1060908 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260530_214851-dgy7zq92/logs/debug-internal.log
+2026-05-30 21:48:51,358 INFO    MainThread:1060908 [wandb_init.py:init():851] calling init triggers
+2026-05-30 21:48:51,358 INFO    MainThread:1060908 [wandb_init.py:init():856] wandb.init called with sweep_config: {}
+config: {'pretrained_model_root': './ckpts', 'pretrained_transformer_version': '480p_t2v', 'post_train_type': 'grpo', 'learning_rate': 1e-05, 'weight_decay': 0.01, 'max_steps': 10000, 'warmup_steps': 500, 'gradient_accumulation_steps': 1, 'max_grad_norm': 1.0, 'use_muon': True, 'num_train_timesteps': 1000, 'train_timestep_shift': 3.0, 'validation_timestep_shift': 5.0, 'snr_type': 'lognorm', 'task_type': 't2v', 'i2v_prob': 0.3, 'enable_fsdp': True, 'enable_gradient_checkpointing': True, 'sp_size': 2, 'dp_replicate': 1, 'batch_size': 2, 'num_workers': 4, 'prefetch_factor': 2, 'num_generations': 4, 'use_same_noise': False, 'output_dir': './outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting', 'checkpoints_directory': 's3://snap-genvid-us-east-2/ywang29/checkpoints_large_model/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting', 'save_interval': 100, 'timer_interval': 25, 'auto_resume_from_timer': True, 'log_interval': 1, 'enable_wandb': True, 'project_name': 'SAGE-GRPO', 'run_name': 'yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting', 'wandb_key': None, 'wandb_username': None, 'wandb_base_url': 'https://api.wandb.ai', 'wandb_entity': None, 'wandb_silent': True, 'dtype': 'bf16', 'master_weight_type': 'fp32', 'seed': 42, 'validation_global_seed': 930, 'validation_interval': 0, 'validation_prompts': None, 'validate_video_length': 121, 'validation_aspect_ratio': '16:9', 'validation_fixed_size': '480x864', 'validation_num_inference_steps': 40, 'validation_negative_prompt': '', 'validation_enable_sr': False, 'validation_prompt_rewrite': False, 'validate_at_step0': False, 'resume_from_checkpoint': None, 'use_lora': False, 'lora_r': 8, 'lora_alpha': 16, 'lora_dropout': 0.0, 'lora_target_modules': None, 'pretrained_lora_path': None, 'train_video_csv': ['/nfs/ywang29/Reward_finetuning/SAGE-GRPO/assets/demo_train_decomposed_binary5.csv'], 'valid_video_csv': 'assets/demo_sample.csv', 'reward_model': 'video_score2', 'reward_config': {'models': {'video_score2': {'weight': 1.0, 'sub_reward': {'VQ': 1.0, 'MQ': 1.0, 'TA': 1.0}}}, 'reward_checkpoint_mode': 'v3', 'reward_gradient_credit': True, 'reward_credit_num_frames': 8, 'reward_credit_metric': 'weighted'}, 'reward_checkpoint_mode': 'v3', 'reward_weights': {'VQ': 1.0, 'MQ': 1.0, 'TA': 1.0}, 'reward_gradient_credit': True, 'reward_credit_num_frames': 8, 'reward_credit_max_questions': 1, 'reward_credit_metric': 'weighted', 'token_credit_source_model': None, 'token_credit_advantage_source': 'token', 'token_credit_remote_url': None, 'remote_reward_url': None, 'vlm_reward_model_path': None, 'vlm_reward_model_family': None, 'vlm_reward_prompt_template': None, 'vlm_reward_num_frames': 8, 'vlm_reward_max_pixels': 65536, 'vlm_reward_max_new_tokens': 128, 'vlm_reward_batch_size': 0, 'vlm_reward_score_type': 'logprob', 'vlm_reward_missing_score': 0.0, 'vlm_reward_video_sample_type': 'uniform', 'vlm_reward_token_credit_max_questions': 8, 'token_credit_loss_mode': 'video_adv', 'token_credit_ratio_mode': 'uniform_weighted_logprob', 'token_credit_shuffle_weights': False, 'enable_grpo_policy_update': True, 'grpo_sampling_steps': 20, 'sample_n_frames': 81, 'eta': 0.5, 'cfg_scale': 6.0, 'neg_prompt': None, 'data_type': 'video', 'model_type': '16164', 'video_fps': 24, 'vae_name': '16164-32c-hy20250605', 'vae_tiling': False, 'infer_flow_shift_video': 5.0, 'video_bucket_hw_base_size': 480, 'video_bucket_hw_bucket_stride': 32, 'multitask_mask_training_type': 'concat', 'vision_num_semantic_tokens': 729, 'vision_states_dim': 1152, 'sde_type': 'sage_grpo', 'mini_batch_size_per_update': 1, 'clip_range': 0.0001, 'adv_clip_max': 5.0, 'timestep_fraction': 1.0, 'random_train_timestep_count': False, 'random_train_timestep_min': 1, 'random_train_timestep_max': 20, 'use_grad_balancing': True, 'enable_timestep_permutation': True, 'reference_mode_offload': True, 'kl_weight': 1e-05, 'kl_coef': 1e-07, 'kl_min_coef': 1e-07, 'use_moving_KL': True, 'update_ref_model_step': 10, 'use_dual_kl': True, 'dual_kl_moving_weight': 1.0, 'dual_kl_step_weight': 0.1, 'kl_compute_mode': 'rollout_phase', 'glyph_byT5_v2': True, 'byt5_max_length': 256, 'video_uncond_p_byt5': 0.0, 'precision': 'bf16', 'vae_precision': 'fp16', 'val_disable_autocast': False, 'debug_grad_flow': False, 'debug_train_diagnostics_interval': 1, '_wandb': {}}
+2026-05-30 21:48:51,358 INFO    MainThread:1060908 [wandb_init.py:init():899] starting backend
+2026-05-30 21:48:51,621 INFO    MainThread:1060908 [wandb_init.py:init():914] sending inform_init request
+2026-05-30 21:48:51,924 INFO    MainThread:1060908 [wandb_init.py:init():919] backend started and connected
+2026-05-30 21:48:51,927 INFO    MainThread:1060908 [wandb_init.py:init():989] updated telemetry
+2026-05-30 21:48:51,976 INFO    MainThread:1060908 [wandb_init.py:init():1012] communicating run to backend with 90.0 second timeout
+2026-05-30 21:48:52,245 INFO    MainThread:1060908 [wandb_init.py:init():1052] run resumed
+2026-05-30 21:48:52,247 INFO    MainThread:1060908 [wandb_init.py:init():1057] starting run threads in backend
+2026-05-30 21:48:52,443 INFO    MainThread:1060908 [wandb_run.py:_console_start():2509] atexit reg
+2026-05-30 21:48:52,443 INFO    MainThread:1060908 [wandb_run.py:_redirect():2359] redirect: wrap_raw
+2026-05-30 21:48:52,443 INFO    MainThread:1060908 [wandb_run.py:_redirect():2428] Wrapping output streams.
+2026-05-30 21:48:52,443 INFO    MainThread:1060908 [wandb_run.py:_redirect():2451] Redirects installed.
+2026-05-30 21:48:52,445 INFO    MainThread:1060908 [wandb_init.py:init():1095] run started, returning control to user process

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260625_180116-dgy7zq92/files/requirements.txt

@@ -0,0 +1,287 @@
+awscli==1.45.30
+boto3==1.43.30
+botocore==1.43.30
+certifi==2026.5.20
+charset-normalizer==3.4.7
+colorama==0.4.6
+docutils==0.19
+idna==3.18
+jmespath==1.1.0
+pip==26.1.2
+pyasn1==0.6.3
+python-dateutil==2.9.0.post0
+PyYAML==6.0.3
+requests==2.34.2
+rsa==4.7.2
+s3transfer==0.18.0
+six==1.17.0
+urllib3==2.7.0
+py-cpuinfo==9.0.0
+hjson==3.1.0
+easydict==1.13
+braceexpand==0.1.7
+antlr4-python3-runtime==4.9.3
+websocket-client==1.9.0
+webdataset==1.0.2
+triton==3.6.0
+torchcodec==0.14.0+cu130
+tifffile==2026.6.1
+tenacity==9.1.4
+smmap==5.0.3
+sentry-sdk==2.63.0
+sentencepiece==0.2.1
+regex==2026.5.9
+pyarrow==24.0.0
+pyaml==26.2.1
+propcache==0.5.2
+opencv-python-headless==4.13.0.92
+omegaconf==2.3.1
+nvidia-nvtx==13.0.85
+ImageIO==2.37.3
+nvidia-nvshmem-cu13==3.4.5
+nvidia-nvjitlink==13.0.88
+av==17.1.0
+nvidia-nccl-cu13==2.28.9
+nvidia-cufile==1.15.1.6
+imageio-ffmpeg==0.6.0
+nvidia-cuda-runtime==13.0.96
+nvidia-cuda-nvrtc==13.0.88
+attrs==26.1.0
+nvidia-cuda-cupti==13.0.85
+nvidia-cublas==13.1.0.3
+multidict==6.7.1
+msgpack==1.2.1
+Mako==1.3.12
+loguru==0.7.3
+lmdb==2.2.1
+lightning-utilities==0.15.3
+lazy-loader==0.5
+greenlet==3.5.2
+future==1.0.0
+ftfy==6.3.1
+frozenlist==1.8.0
+frozendict==2.4.7
+docstring_parser==0.18.0
+decord==0.6.0
+cuda-pathfinder==1.5.5
+aiohappyeyeballs==2.6.2
+yarl==1.24.2
+typed-argument-parser==1.12.0
+SQLAlchemy==2.0.51
+scikit-image==0.26.0
+qwen-vl-utils==0.0.14
+typeguard==4.5.2
+nvidia-cusparse==12.6.3.3
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.15.1.9
+huggingface_hub==0.36.2
+gitdb==4.0.12
+ffmpeg-python==0.2.0
+einx==0.4.3
+cupy-cuda12x==14.1.1
+cuda-bindings==13.0.3
+cattrs==26.1.0
+aiosignal==1.4.0
+tokenizers==0.21.4
+termcolor==3.3.0
+nvidia-cusolver==12.0.4.66
+mediapy==1.2.6
+ipdb==0.13.13
+GitPython==3.1.50
+diffusers==0.38.0
+coremltools==9.0
+alembic==1.18.5
+aiohttp==3.14.1
+wandb==0.28.0
+torch==2.10.0+cu130
+dataset==2.0.0
+dashscope==1.26.0
+torchvision==0.25.0+cu130
+torchmetrics==1.9.0
+sniffio==1.3.1
+torchdata==0.8.0
+tqdm==4.67.3
+torchaudio==2.10.0+cu130
+deepspeed==0.19.2
+timm==1.0.27
+pytorch-lightning==2.6.5
+peft==0.19.1
+lpips==0.1.4
+lightning==2.6.5
+flash_attn==2.8.3+cu130torch2.10
+xxhash==3.7.1
+shortuuid==1.0.13
+setuptools==82.0.0
+safetensors==0.7.0
+pillow==12.1.1
+numpy==2.4.2
+jiter==0.15.0
+fsspec==2026.2.0
+distro==1.9.0
+dill==0.4.1
+tyro==1.0.15
+tiktoken==0.13.0
+multiprocess==0.70.19
+modelscope==1.36.0
+fire==0.7.0
+openai==2.8.0
+datasets==4.8.4
+trl==0.8.6
+angelslim==0.2.2
+transformers==4.53.2
+black==26.5.1
+pathspec==1.1.1
+pytokens==0.4.1
+importlib_metadata==9.0.0
+ml_dtypes==0.5.4
+onnx==1.22.0
+onnx-ir==0.2.1
+onnxscript==0.7.0
+transformer_engine==2.9.0+70f53666
+zipp==4.1.0
+nvidia-mathdx==25.6.0
+einops==0.8.2
+absl-py==2.4.0
+accelerate==1.14.0
+annotated-doc==0.0.4
+annotated-types==0.7.0
+anyio==4.14.0
+apsw==3.53.2.0
+apswutils==0.1.2
+beartype==0.22.9
+beautifulsoup4==4.15.0
+blis==1.2.1
+catalogue==2.0.10
+cloudpathlib==0.24.0
+cloudpickle==3.1.2
+confection==0.1.5
+contourpy==1.3.3
+cycler==0.12.1
+cymem==2.0.13
+fastai==2.8.7
+fastcore==1.13.6
+fastdownload==0.0.7
+fastlite==0.2.4
+fastprogress==1.1.6
+fasttransform==0.0.2
+fonttools==4.63.0
+grpcio==1.81.1
+h11==0.16.0
+hf-xet==1.5.1
+httpcore==1.0.9
+httpcore2==2.4.0
+httptools==0.8.0
+httpx==0.28.1
+httpx2==2.4.0
+itsdangerous==2.2.0
+joblib==1.5.3
+kiwisolver==1.5.0
+langcodes==3.5.1
+Markdown==3.10.2
+markdown-it-py==4.2.0
+matplotlib==3.11.0
+mdurl==0.1.2
+mpmath==1.3.0
+murmurhash==1.0.15
+mypy_extensions==1.1.0
+narwhals==2.22.1
+networkx==3.6.1
+nvidia-curand==10.4.0.35
+nvidia-cusparselt-cu13==0.8.0
+oauthlib==3.3.1
+pandas==3.0.3
+plum-dispatch==2.9.0
+preshed==3.0.13
+protobuf==7.35.1
+pydantic==2.13.4
+pydantic_core==2.46.4
+pyparsing==3.3.2
+pyre-extensions==0.0.32
+python-dotenv==1.2.2
+python-fasthtml==0.14.3
+python-multipart==0.0.32
+rich==15.0.0
+s3torchconnector==1.5.0
+s3torchconnectorclient==1.5.0
+scikit-learn==1.9.0
+shellingham==1.5.4
+smart_open==7.6.1
+soupsieve==2.8.4
+spacy==3.8.7
+spacy-legacy==3.0.12
+spacy-loggers==1.0.5
+srsly==2.5.3
+starlette==1.3.1
+sympy==1.14.0
+tabulate==0.10.0
+tensorboard==2.20.0
+tensorboard-data-server==0.7.2
+thinc==8.3.4
+threadpoolctl==3.6.0
+torchtnt==0.2.4
+truststore==0.10.4
+typer==0.25.1
+typer-slim==0.24.0
+typing-inspect==0.9.0
+typing-inspection==0.4.2
+uvicorn==0.49.0
+uvloop==0.22.1
+wasabi==1.1.3
+watchfiles==1.2.0
+weasel==0.4.3
+websockets==16.0
+Werkzeug==3.1.8
+wrapt==2.2.2
+asttokens==3.0.1
+cffi==2.0.0
+click==8.4.1
+comm==0.2.3
+cryptography==49.0.0
+Cython==3.2.5
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+filelock==3.29.4
+h5py==3.16.0
+intel-cmplr-lib-ur==2026.0.0
+intel-openmp==2026.0.0
+ipykernel==7.3.0
+ipython==9.14.1
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+Jinja2==3.1.6
+jupyter_client==8.9.1
+jupyter_core==5.9.1
+MarkupSafe==3.0.3
+matplotlib-inline==0.2.2
+mkl==2026.0.0
+mkl-include==2026.0.0
+mpi4py==4.1.2
+nest-asyncio2==1.7.2
+ninja==1.13.0
+onemkl-license==2026.0.0
+opencv-python==4.11.0.86
+packaging==26.2
+parso==0.8.7
+pexpect==4.9.0
+platformdirs==4.10.0
+prompt_toolkit==3.0.52
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+pybind11==3.0.4
+pycparser==3.0
+Pygments==2.20.0
+pyOpenSSL==26.3.0
+pyzmq==27.1.0
+scipy==1.18.0
+stack-data==0.6.3
+tbb==2023.0.0
+tcmlib==1.5.0
+tornado==6.5.7
+traitlets==5.15.1
+typing_extensions==4.15.0
+tzdata==2026.2
+umf==1.1.0
+wheel==0.47.0
+wcwidth==0.8.1

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260625_180116-dgy7zq92/files/wandb-metadata.json

@@ -0,0 +1,167 @@
+{
+  "os": "Linux-6.1.163-186.299.amzn2023.x86_64-x86_64-with-glibc2.35",
+  "python": "CPython 3.12.10",
+  "startedAt": "2026-06-25T18:01:16.324631Z",
+  "args": [
+    "--pretrained_model_root",
+    "./ckpts",
+    "--train_video_csv",
+    "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/assets/demo_train_decomposed_binary5.csv",
+    "--reward_model",
+    "video_score2",
+    "--reward_weights",
+    "{\"VQ\":1.0,\"MQ\":1.0,\"TA\":1.0}",
+    "--reward_gradient_credit",
+    "True",
+    "--reward_credit_num_frames",
+    "8",
+    "--reward_credit_max_questions",
+    "1",
+    "--reward_credit_metric",
+    "weighted",
+    "--token_credit_loss_mode",
+    "video_adv",
+    "--token_credit_ratio_mode",
+    "uniform_weighted_logprob",
+    "--auto_resume_from_timer",
+    "True",
+    "--learning_rate",
+    "1e-5",
+    "--batch_size",
+    "2",
+    "--num_generations",
+    "4",
+    "--max_steps",
+    "10000",
+    "--output_dir",
+    "./outputs",
+    "--enable_wandb",
+    "True",
+    "--project_name",
+    "SAGE-GRPO",
+    "--run_name",
+    "yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting",
+    "--enable_fsdp",
+    "--enable_gradient_checkpointing",
+    "--sp_size",
+    "2",
+    "--reward_checkpoint_mode",
+    "v3",
+    "--validation_interval",
+    "0",
+    "--validate_at_step0",
+    "False",
+    "--validate_video_length",
+    "121",
+    "--validation_timestep_shift",
+    "5.0",
+    "--use_grad_balancing",
+    "True",
+    "--enable_timestep_permutation",
+    "True",
+    "--sde_type",
+    "sage_grpo",
+    "--kl_weight",
+    "1e-5",
+    "--kl_coef",
+    "1e-7",
+    "--use_moving_KL",
+    "True",
+    "--update_ref_model_step",
+    "10",
+    "--use_dual_kl",
+    "True",
+    "--dual_kl_moving_weight",
+    "1.0",
+    "--dual_kl_step_weight",
+    "0.1",
+    "--reference_mode_offload",
+    "True",
+    "--checkpoints_directory",
+    "s3://snap-genvid-us-east-2/ywang29/checkpoints_large_model/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting"
+  ],
+  "program": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/post_train.py",
+  "codePath": "post_train.py",
+  "codePathLocal": "post_train.py",
+  "git": {
+    "remote": "https://github.com/Tencent-Hunyuan/SAGE-GRPO.git",
+    "commit": "779776e42732e0262b3f2844dd206d9d8f043b2f"
+  },
+  "root": "./outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb",
+  "host": "hunyuan-sage-vs2-tcva-nw-worker-0",
+  "username": "yi-fan-wang1216",
+  "executable": "/usr/local/bin/python",
+  "cpu_count": 48,
+  "cpu_count_logical": 96,
+  "gpu": "NVIDIA A100-SXM4-80GB",
+  "gpu_count": 8,
+  "disk": {
+    "/": {
+      "total": "7516777873408",
+      "used": "729981288448"
+    }
+  },
+  "memory": {
+    "total": "1204539252736"
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-f6eafbbf-b72f-1725-61a0-5da7fda21fe2"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-8faea6eb-ec0a-0561-c1a7-608821eb24f5"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-6d3845d3-5be7-3130-93f6-aa8bc1fc8f9b"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-a80bc0c5-98ed-a833-ec83-51e290f6ac9c"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-a8dcbb50-ac55-1c6b-f4a3-66a2e17a2e9f"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-a6632210-2f96-54eb-7c24-6689d8c3ab90"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-e2324dd0-03ea-7053-4212-42facf1acdf7"
+    },
+    {
+      "name": "NVIDIA A100-SXM4-80GB",
+      "memoryTotal": "85899345920",
+      "cudaCores": 6912,
+      "architecture": "Ampere",
+      "uuid": "GPU-20e27620-2f39-65f1-ff0c-215b32505ff5"
+    }
+  ],
+  "cudaVersion": "13.0",
+  "writerId": "2cdf2x7re0osw6giy31rnz0o7wfy0p90"
+}

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260625_180116-dgy7zq92/logs/debug.log

@@ -0,0 +1,19 @@
+2026-06-25 18:01:16,347 INFO    MainThread:1812 [wandb_setup.py:_flush():81] Current SDK version is 0.28.0
+2026-06-25 18:01:16,347 INFO    MainThread:1812 [wandb_setup.py:_flush():81] Configure stats pid to 1812
+2026-06-25 18:01:16,348 INFO    MainThread:1812 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-06-25 18:01:16,348 INFO    MainThread:1812 [wandb_init.py:setup_run_log_directory():725] Logging user logs to outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260625_180116-dgy7zq92/logs/debug.log
+2026-06-25 18:01:16,348 INFO    MainThread:1812 [wandb_init.py:setup_run_log_directory():726] Logging internal logs to outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/wandb/wandb/run-20260625_180116-dgy7zq92/logs/debug-internal.log
+2026-06-25 18:01:16,348 INFO    MainThread:1812 [wandb_init.py:init():768] calling init triggers
+2026-06-25 18:01:16,348 INFO    MainThread:1812 [wandb_init.py:init():773] wandb.init called with sweep_config: {}
+config: {'pretrained_model_root': './ckpts', 'pretrained_transformer_version': '480p_t2v', 'post_train_type': 'grpo', 'learning_rate': 1e-05, 'weight_decay': 0.01, 'max_steps': 10000, 'warmup_steps': 500, 'gradient_accumulation_steps': 1, 'max_grad_norm': 1.0, 'use_muon': True, 'num_train_timesteps': 1000, 'train_timestep_shift': 3.0, 'validation_timestep_shift': 5.0, 'snr_type': 'lognorm', 'task_type': 't2v', 'i2v_prob': 0.3, 'enable_fsdp': True, 'enable_gradient_checkpointing': True, 'sp_size': 2, 'dp_replicate': 1, 'batch_size': 2, 'num_workers': 4, 'prefetch_factor': 2, 'num_generations': 4, 'use_same_noise': False, 'output_dir': './outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting', 'checkpoints_directory': 's3://snap-genvid-us-east-2/ywang29/checkpoints_large_model/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting', 'save_interval': 100, 'timer_interval': 25, 'auto_resume_from_timer': True, 'log_interval': 1, 'enable_wandb': True, 'project_name': 'SAGE-GRPO', 'run_name': 'yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting', 'wandb_key': None, 'wandb_username': None, 'wandb_base_url': 'https://api.wandb.ai', 'wandb_entity': None, 'wandb_silent': True, 'dtype': 'bf16', 'master_weight_type': 'fp32', 'seed': 42, 'validation_global_seed': 930, 'validation_interval': 0, 'validation_prompts': None, 'validate_video_length': 121, 'validation_aspect_ratio': '16:9', 'validation_fixed_size': '480x864', 'validation_num_inference_steps': 40, 'validation_negative_prompt': '', 'validation_enable_sr': False, 'validation_prompt_rewrite': False, 'validate_at_step0': False, 'resume_from_checkpoint': None, 'use_lora': False, 'lora_r': 8, 'lora_alpha': 16, 'lora_dropout': 0.0, 'lora_target_modules': None, 'pretrained_lora_path': None, 'train_video_csv': ['/nfs/ywang29/Reward_finetuning/SAGE-GRPO/assets/demo_train_decomposed_binary5.csv'], 'valid_video_csv': 'assets/demo_sample.csv', 'reward_model': 'video_score2', 'reward_config': {'models': {'video_score2': {'weight': 1.0, 'sub_reward': {'VQ': 1.0, 'MQ': 1.0, 'TA': 1.0}}}, 'reward_checkpoint_mode': 'v3', 'reward_gradient_credit': True, 'reward_credit_num_frames': 8, 'reward_credit_metric': 'weighted'}, 'reward_checkpoint_mode': 'v3', 'reward_weights': {'VQ': 1.0, 'MQ': 1.0, 'TA': 1.0}, 'reward_gradient_credit': True, 'reward_credit_num_frames': 8, 'reward_credit_max_questions': 1, 'reward_credit_metric': 'weighted', 'token_credit_source_model': None, 'token_credit_advantage_source': 'token', 'token_credit_remote_url': None, 'token_credit_vlm_model_path': None, 'token_credit_vlm_model_family': None, 'token_credit_vlm_prompt_template': None, 'remote_reward_url': None, 'vlm_reward_model_path': None, 'vlm_reward_model_family': None, 'vlm_reward_prompt_template': None, 'vlm_reward_num_frames': 8, 'vlm_reward_max_pixels': 65536, 'vlm_reward_max_new_tokens': 128, 'vlm_reward_batch_size': 0, 'vlm_reward_score_type': 'logprob', 'vlm_reward_missing_score': 0.0, 'vlm_reward_video_sample_type': 'uniform', 'vlm_reward_token_credit_max_questions': 8, 'token_credit_loss_mode': 'video_adv', 'token_credit_ratio_mode': 'uniform_weighted_logprob', 'token_credit_shuffle_weights': False, 'token_credit_spatial_granularity': 'frame', 'token_credit_spatial_bins': 1, 'token_credit_spatial_window': 1, 'token_credit_chunk_size': 1, 'enable_grpo_policy_update': True, 'grpo_sampling_steps': 20, 'sample_n_frames': 81, 'eta': 0.5, 'cfg_scale': 6.0, 'neg_prompt': None, 'data_type': 'video', 'model_type': '16164', 'video_fps': 24, 'vae_name': '16164-32c-hy20250605', 'vae_tiling': False, 'infer_flow_shift_video': 5.0, 'video_bucket_hw_base_size': 480, 'video_bucket_hw_bucket_stride': 32, 'multitask_mask_training_type': 'concat', 'vision_num_semantic_tokens': 729, 'vision_states_dim': 1152, 'sde_type': 'sage_grpo', 'mini_batch_size_per_update': 1, 'clip_range': 0.0001, 'adv_clip_max': 5.0, 'timestep_fraction': 1.0, 'random_train_timestep_count': False, 'random_train_timestep_min': 1, 'random_train_timestep_max': 20, 'use_grad_balancing': True, 'enable_timestep_permutation': True, 'reference_mode_offload': True, 'kl_weight': 1e-05, 'kl_coef': 1e-07, 'kl_min_coef': 1e-07, 'use_moving_KL': True, 'update_ref_model_step': 10, 'use_dual_kl': True, 'dual_kl_moving_weight': 1.0, 'dual_kl_step_weight': 0.1, 'kl_compute_mode': 'rollout_phase', 'glyph_byT5_v2': True, 'byt5_max_length': 256, 'video_uncond_p_byt5': 0.0, 'precision': 'bf16', 'vae_precision': 'fp16', 'val_disable_autocast': False, 'debug_grad_flow': False, 'debug_train_diagnostics_interval': 1, '_wandb': {}}
+2026-06-25 18:01:16,348 INFO    MainThread:1812 [wandb_init.py:init():816] starting backend
+2026-06-25 18:01:16,613 INFO    MainThread:1812 [wandb_init.py:init():831] sending inform_init request
+2026-06-25 18:01:16,944 INFO    MainThread:1812 [wandb_init.py:init():836] backend started and connected
+2026-06-25 18:01:16,947 INFO    MainThread:1812 [wandb_init.py:init():906] updated telemetry
+2026-06-25 18:01:17,001 INFO    MainThread:1812 [wandb_init.py:init():929] communicating run to backend with 90.0 second timeout
+2026-06-25 18:01:17,475 INFO    MainThread:1812 [wandb_init.py:init():974] starting run threads in backend
+2026-06-25 18:01:17,684 INFO    MainThread:1812 [wandb_run.py:_console_start():2523] atexit reg
+2026-06-25 18:01:17,684 INFO    MainThread:1812 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-06-25 18:01:17,684 INFO    MainThread:1812 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-06-25 18:01:17,684 INFO    MainThread:1812 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-06-25 18:01:17,686 INFO    MainThread:1812 [wandb_init.py:init():1012] run started, returning control to user process