LLDDSS commited on
Commit
d9fd354
·
verified ·
1 Parent(s): 39a33f9

Upload folder using huggingface_hub (part 5)

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/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics.json +25 -0
  2. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard0.json +26 -0
  3. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard1.json +26 -0
  4. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard2.json +26 -0
  5. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard3.json +26 -0
  6. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions.jsonl +0 -0
  7. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard0.jsonl +0 -0
  8. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard1.jsonl +0 -0
  9. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard2.jsonl +0 -0
  10. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard3.jsonl +0 -0
  11. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics.json +25 -0
  12. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard0.json +26 -0
  13. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard1.json +26 -0
  14. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard2.json +26 -0
  15. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard3.json +26 -0
  16. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions.jsonl +0 -0
  17. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard0.jsonl +0 -0
  18. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard1.jsonl +0 -0
  19. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard2.jsonl +0 -0
  20. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard3.jsonl +0 -0
  21. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics.json +25 -0
  22. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard0.json +26 -0
  23. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard1.json +26 -0
  24. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard2.json +26 -0
  25. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard3.json +26 -0
  26. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions.jsonl +0 -0
  27. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard0.jsonl +0 -0
  28. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard1.jsonl +0 -0
  29. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard2.jsonl +0 -0
  30. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard3.jsonl +0 -0
  31. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics.json +25 -0
  32. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard0.json +26 -0
  33. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard1.json +26 -0
  34. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard2.json +26 -0
  35. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard3.json +26 -0
  36. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions.jsonl +0 -0
  37. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard0.jsonl +0 -0
  38. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard1.jsonl +0 -0
  39. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard2.jsonl +0 -0
  40. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard3.jsonl +0 -0
  41. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics.json +25 -0
  42. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard0.json +26 -0
  43. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard1.json +26 -0
  44. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard2.json +26 -0
  45. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard3.json +26 -0
  46. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions.jsonl +0 -0
  47. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard0.jsonl +0 -0
  48. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard1.jsonl +0 -0
  49. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard2.jsonl +0 -0
  50. result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard3.jsonl +0 -0
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.00",
7
+ "scoring": "likelihood",
8
+ "num_samples": 764,
9
+ "accuracy": 0.418848167539267,
10
+ "num_correct": 320,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.50293445587158,
24
+ "samples_per_second": 20.371739200807674
25
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.00",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.4083769633507853,
10
+ "num_correct": 78,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.55298376083374,
25
+ "samples_per_second": 5.2252916273460315
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.00",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.450261780104712,
10
+ "num_correct": 86,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.29963827133179,
25
+ "samples_per_second": 5.261760422302755
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.00",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.42408376963350786,
10
+ "num_correct": 81,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.50293445587158,
25
+ "samples_per_second": 5.092934800201919
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.00",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.39267015706806285,
10
+ "num_correct": 75,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.83789014816284,
25
+ "samples_per_second": 5.184878917652276
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.00/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.05",
7
+ "scoring": "likelihood",
8
+ "num_samples": 764,
9
+ "accuracy": 0.42277486910994766,
10
+ "num_correct": 323,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.67427349090576,
24
+ "samples_per_second": 20.27909045636734
25
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.05",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.41361256544502617,
10
+ "num_correct": 79,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.37861680984497,
25
+ "samples_per_second": 5.250337059222949
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.05",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.4397905759162304,
10
+ "num_correct": 84,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.25076460838318,
25
+ "samples_per_second": 5.268854383166039
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.05",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.42408376963350786,
10
+ "num_correct": 81,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.67427349090576,
25
+ "samples_per_second": 5.069772614091835
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.05",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.41361256544502617,
10
+ "num_correct": 79,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.60825705528259,
25
+ "samples_per_second": 5.217402175459172
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.05/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.10",
7
+ "scoring": "likelihood",
8
+ "num_samples": 764,
9
+ "accuracy": 0.4293193717277487,
10
+ "num_correct": 328,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 38.091694355010986,
24
+ "samples_per_second": 20.056865753452506
25
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.10",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.4397905759162304,
10
+ "num_correct": 84,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.95055961608887,
25
+ "samples_per_second": 5.1690692098973114
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.10",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.43455497382198954,
10
+ "num_correct": 83,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.7716019153595,
25
+ "samples_per_second": 5.194225708187581
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.10",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.418848167539267,
10
+ "num_correct": 80,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 38.091694355010986,
25
+ "samples_per_second": 5.0142164383631265
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.10",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.42408376963350786,
10
+ "num_correct": 81,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.881108045578,
25
+ "samples_per_second": 5.178803190076624
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.10/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.15",
7
+ "scoring": "likelihood",
8
+ "num_samples": 764,
9
+ "accuracy": 0.43717277486910994,
10
+ "num_correct": 334,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.60738945007324,
24
+ "samples_per_second": 20.315156440578516
25
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.15",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.450261780104712,
10
+ "num_correct": 86,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.62264108657837,
25
+ "samples_per_second": 5.215352971088657
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.15",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.4293193717277487,
10
+ "num_correct": 82,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.70540809631348,
25
+ "samples_per_second": 5.203592873802789
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.15",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.4293193717277487,
10
+ "num_correct": 82,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.60738945007324,
25
+ "samples_per_second": 5.078789110144629
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.15",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.4397905759162304,
10
+ "num_correct": 84,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.60187005996704,
25
+ "samples_per_second": 5.218312607718492
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.15/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.20",
7
+ "scoring": "likelihood",
8
+ "num_samples": 764,
9
+ "accuracy": 0.4476439790575916,
10
+ "num_correct": 342,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.8256413936615,
24
+ "samples_per_second": 20.197939065959226
25
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard0.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.20",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.45549738219895286,
10
+ "num_correct": 87,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.1693811416626,
25
+ "samples_per_second": 5.138638151441025
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard1.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.20",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.44502617801047123,
10
+ "num_correct": 85,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 36.689135789871216,
25
+ "samples_per_second": 5.205900762937279
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard2.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.20",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.4397905759162304,
10
+ "num_correct": 84,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.8256413936615,
25
+ "samples_per_second": 5.049484766489806
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/metrics_shard3.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "Qwen/Qwen3.5-4B-Base",
3
+ "dataset": "youtube",
4
+ "level": 2,
5
+ "task": "youtube_level2",
6
+ "baseline": "0.20",
7
+ "scoring": "likelihood",
8
+ "num_samples": 191,
9
+ "accuracy": 0.450261780104712,
10
+ "num_correct": 86,
11
+ "num_unparsed": 0,
12
+ "unparsed_rate": 0.0,
13
+ "answer_instruction": "",
14
+ "num_choices": 4,
15
+ "has_chat_template": false,
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": 37.18783354759216,
25
+ "samples_per_second": 5.136088386422469
26
+ }
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard0.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard1.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard2.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
result/_archive_l2_v1/Qwen_Qwen3.5-4B-Base/youtube_level2/baseline_0.20/predictions_shard3.jsonl ADDED
The diff for this file is too large to render. See raw diff