LLDDSS commited on
Commit
d67b395
·
verified ·
1 Parent(s): 18104f9

Upload folder using huggingface_hub (part 2)

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/metrics_shard3.json +26 -0
  2. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions.jsonl +0 -0
  3. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard0.jsonl +0 -0
  4. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard1.jsonl +0 -0
  5. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard2.jsonl +0 -0
  6. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard3.jsonl +0 -0
  7. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics.json +25 -0
  8. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard0.json +26 -0
  9. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard1.json +26 -0
  10. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard2.json +26 -0
  11. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard3.json +26 -0
  12. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions.jsonl +0 -0
  13. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard0.jsonl +0 -0
  14. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard1.jsonl +0 -0
  15. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard2.jsonl +0 -0
  16. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard3.jsonl +0 -0
  17. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics.json +25 -0
  18. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard0.json +26 -0
  19. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard1.json +26 -0
  20. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard2.json +26 -0
  21. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard3.json +26 -0
  22. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions.jsonl +0 -0
  23. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard0.jsonl +0 -0
  24. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard1.jsonl +0 -0
  25. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard2.jsonl +0 -0
  26. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard3.jsonl +0 -0
  27. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics.json +25 -0
  28. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard0.json +26 -0
  29. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard1.json +26 -0
  30. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard2.json +26 -0
  31. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard3.json +26 -0
  32. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions.jsonl +0 -0
  33. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard0.jsonl +0 -0
  34. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard1.jsonl +0 -0
  35. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard2.jsonl +0 -0
  36. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard3.jsonl +0 -0
  37. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics.json +25 -0
  38. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard0.json +26 -0
  39. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard1.json +26 -0
  40. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard2.json +26 -0
  41. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard3.json +26 -0
  42. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions.jsonl +0 -0
  43. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard0.jsonl +0 -0
  44. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard1.jsonl +0 -0
  45. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard2.jsonl +0 -0
  46. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard3.jsonl +0 -0
  47. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics.json +25 -0
  48. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics_shard0.json +26 -0
  49. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics_shard1.json +26 -0
  50. result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics_shard2.json +26 -0
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.00",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.24401913875598086,
10
+ "num_correct": 51,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 3,
24
+ "elapsed_seconds": 40.40080738067627,
25
+ "samples_per_second": 5.173163942757363
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.00/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.05",
7
+ "scoring": "generate",
8
+ "num_samples": 836,
9
+ "accuracy": 0.2452153110047847,
10
+ "num_correct": 205,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "elapsed_seconds": 41.78576159477234,
24
+ "samples_per_second": 20.006814955470116
25
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.05",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.3014354066985646,
10
+ "num_correct": 63,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 0,
24
+ "elapsed_seconds": 40.12726616859436,
25
+ "samples_per_second": 5.208428581251669
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.05",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2631578947368421,
10
+ "num_correct": 55,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 1,
24
+ "elapsed_seconds": 41.28731918334961,
25
+ "samples_per_second": 5.062086958755262
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.05",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.19138755980861244,
10
+ "num_correct": 40,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 2,
24
+ "elapsed_seconds": 41.78576159477234,
25
+ "samples_per_second": 5.001703738867529
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.05",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.22488038277511962,
10
+ "num_correct": 47,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 3,
24
+ "elapsed_seconds": 41.28069281578064,
25
+ "samples_per_second": 5.062899523820594
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.05/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.10",
7
+ "scoring": "generate",
8
+ "num_samples": 836,
9
+ "accuracy": 0.23923444976076555,
10
+ "num_correct": 200,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "elapsed_seconds": 41.29511094093323,
24
+ "samples_per_second": 20.24452728062116
25
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.10",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.291866028708134,
10
+ "num_correct": 61,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 0,
24
+ "elapsed_seconds": 40.01466393470764,
25
+ "samples_per_second": 5.223085225482027
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.10",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2535885167464115,
10
+ "num_correct": 53,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 1,
24
+ "elapsed_seconds": 40.32780385017395,
25
+ "samples_per_second": 5.182528678637642
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.10",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.18181818181818182,
10
+ "num_correct": 38,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 2,
24
+ "elapsed_seconds": 41.29511094093323,
25
+ "samples_per_second": 5.06113182015529
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.10",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.22966507177033493,
10
+ "num_correct": 48,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 3,
24
+ "elapsed_seconds": 40.534538984298706,
25
+ "samples_per_second": 5.156096633563722
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.10/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.15",
7
+ "scoring": "generate",
8
+ "num_samples": 836,
9
+ "accuracy": 0.25239234449760767,
10
+ "num_correct": 211,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "elapsed_seconds": 41.511862993240356,
24
+ "samples_per_second": 20.13882152521392
25
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.15",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2727272727272727,
10
+ "num_correct": 57,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 0,
24
+ "elapsed_seconds": 40.5939257144928,
25
+ "samples_per_second": 5.148553541481776
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.15",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2727272727272727,
10
+ "num_correct": 57,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 1,
24
+ "elapsed_seconds": 40.53896141052246,
25
+ "samples_per_second": 5.15553415104885
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.15",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.22488038277511962,
10
+ "num_correct": 47,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 2,
24
+ "elapsed_seconds": 41.511862993240356,
25
+ "samples_per_second": 5.03470538130348
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.15",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.23923444976076555,
10
+ "num_correct": 50,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 3,
24
+ "elapsed_seconds": 40.63222599029541,
25
+ "samples_per_second": 5.143700471884497
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.15/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.20",
7
+ "scoring": "generate",
8
+ "num_samples": 836,
9
+ "accuracy": 0.2595693779904306,
10
+ "num_correct": 217,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "elapsed_seconds": 41.89776062965393,
24
+ "samples_per_second": 19.953333720855362
25
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.20",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2535885167464115,
10
+ "num_correct": 53,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 0,
24
+ "elapsed_seconds": 40.195486545562744,
25
+ "samples_per_second": 5.199588758879495
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.20",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2966507177033493,
10
+ "num_correct": 62,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 1,
24
+ "elapsed_seconds": 40.00703024864197,
25
+ "samples_per_second": 5.2240818351443234
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.20",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.23923444976076555,
10
+ "num_correct": 50,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 2,
24
+ "elapsed_seconds": 41.89776062965393,
25
+ "samples_per_second": 4.9883334302138405
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.20",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.24880382775119617,
10
+ "num_correct": 52,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 3,
24
+ "elapsed_seconds": 40.14028787612915,
25
+ "samples_per_second": 5.20673894130912
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.20/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.25",
7
+ "scoring": "generate",
8
+ "num_samples": 836,
9
+ "accuracy": 0.26555023923444976,
10
+ "num_correct": 222,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "elapsed_seconds": 42.45227670669556,
24
+ "samples_per_second": 19.69270118952528
25
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.25",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.291866028708134,
10
+ "num_correct": 61,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 0,
24
+ "elapsed_seconds": 40.7334680557251,
25
+ "samples_per_second": 5.1309159267774405
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.25",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2727272727272727,
10
+ "num_correct": 57,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 1,
24
+ "elapsed_seconds": 40.81743621826172,
25
+ "samples_per_second": 5.120360790972301
26
+ }
result/Qwen_Qwen2.5-VL-3B-Instruct/youtube_level3/baseline_0.25/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen2.5-VL-3B-Instruct",
3
+ "dataset": "youtube",
4
+ "level": 3,
5
+ "task": "youtube_level3",
6
+ "baseline": "0.25",
7
+ "scoring": "generate",
8
+ "num_samples": 209,
9
+ "accuracy": 0.2535885167464115,
10
+ "num_correct": 53,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": true,
16
+ "enable_thinking": false,
17
+ "batch_size": 8,
18
+ "max_new_tokens": 16,
19
+ "image_min_pixels": 401408,
20
+ "image_max_pixels": 1003520,
21
+ "dtype": "bf16",
22
+ "num_shards": 4,
23
+ "shard_id": 2,
24
+ "elapsed_seconds": 42.45227670669556,
25
+ "samples_per_second": 4.92317529738132
26
+ }