Hugging Face's logo

mahPII
/
models_metrics_499

Safetensors
Model card Files
xet
 Community
mah04 commited on
Commit
ec22a82
·
verified ·
1 Parent(s): f240be6

Recompute metrics batch 8

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. Llama3B/trainingExperiments/adapter/EnglishEasy/350/v2/experiments/adapter_20260404_190652/metrics_new.json +24 -0
  2. Llama3B/trainingExperiments/adapter/EnglishEasy/350/v2/experiments/adapter_20260404_190652/results_with_metrics_new.csv +0 -0
  3. Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_20260404_144425/metrics_new.json +24 -0
  4. Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_20260404_144425/results_with_metrics_new.csv +0 -0
  5. Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_test_20260420_124234/metrics_test_new.json +24 -0
  6. Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_test_20260420_124234/results_test_with_metrics_new.csv +0 -0
  7. Llama3B/trainingExperiments/adapter/EnglishEasy/50/v2/experiments/adapter_20260404_151236/metrics_new.json +24 -0
  8. Llama3B/trainingExperiments/adapter/EnglishEasy/50/v2/experiments/adapter_20260404_151236/results_with_metrics_new.csv +0 -0
  9. Llama3B/trainingExperiments/adapter/EnglishHard/1000/experiments/adapter_20260405_211432/metrics_new.json +24 -0
  10. Llama3B/trainingExperiments/adapter/EnglishHard/1000/experiments/adapter_20260405_211432/results_with_metrics_new.csv +0 -0
  11. Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_20260406_013129/metrics_new.json +24 -0
  12. Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_20260406_013129/results_with_metrics_new.csv +0 -0
  13. Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_test_20260420_235923/metrics_test_new.json +24 -0
  14. Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_test_20260420_235923/results_test_with_metrics_new.csv +0 -0
  15. Llama3B/trainingExperiments/adapter/EnglishHard/350/experiments/adapter_20260405_133314/metrics_new.json +24 -0
  16. Llama3B/trainingExperiments/adapter/EnglishHard/350/experiments/adapter_20260405_133314/results_with_metrics_new.csv +0 -0
  17. Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_20260405_155859/metrics_new.json +24 -0
  18. Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_20260405_155859/results_with_metrics_new.csv +0 -0
  19. Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_test_20260420_215936/metrics_test_new.json +24 -0
  20. Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_test_20260420_215936/results_test_with_metrics_new.csv +0 -0
  21. Llama3B/trainingExperiments/adapter/EnglishHard/50/experiments/adapter_20260405_120144/metrics_new.json +24 -0
  22. Llama3B/trainingExperiments/adapter/EnglishHard/50/experiments/adapter_20260405_120144/results_with_metrics_new.csv +0 -0
  23. Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_20260405_123002/metrics_new.json +24 -0
  24. Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_20260405_123002/results_with_metrics_new.csv +0 -0
  25. Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_test_20260420_213827/metrics_test_new.json +24 -0
  26. Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_test_20260420_213827/results_test_with_metrics_new.csv +0 -0
  27. Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_20260408_144903/metrics_new.json +24 -0
  28. Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_20260408_144903/results_with_metrics_new.csv +0 -0
  29. Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_test_20260421_090302/metrics_test_new.json +24 -0
  30. Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_test_20260421_090302/results_test_with_metrics_new.csv +0 -0
  31. Llama3B/trainingExperiments/adapter/Mixed/1000/v2/experiments/adapter_20260409_001233/metrics_new.json +24 -0
  32. Llama3B/trainingExperiments/adapter/Mixed/1000/v2/experiments/adapter_20260409_001233/results_with_metrics_new.csv +0 -0
  33. Llama3B/trainingExperiments/adapter/Mixed/350/experiments/adapter_20260408_065136/metrics_new.json +24 -0
  34. Llama3B/trainingExperiments/adapter/Mixed/350/experiments/adapter_20260408_065136/results_with_metrics_new.csv +0 -0
  35. Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_20260408_093702/metrics_new.json +24 -0
  36. Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_20260408_093702/results_with_metrics_new.csv +0 -0
  37. Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_test_20260421_070514/metrics_test_new.json +24 -0
  38. Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_test_20260421_070514/results_test_with_metrics_new.csv +0 -0
  39. Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_20260408_051945/metrics_new.json +24 -0
  40. Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_20260408_051945/results_with_metrics_new.csv +0 -0
  41. Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_test_20260421_064404/metrics_test_new.json +24 -0
  42. Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_test_20260421_064404/results_test_with_metrics_new.csv +0 -0
  43. Llama3B/trainingExperiments/adapter/Mixed/50/v2/experiments/adapter_20260408_054820/metrics_new.json +24 -0
  44. Llama3B/trainingExperiments/adapter/Mixed/50/v2/experiments/adapter_20260408_054820/results_with_metrics_new.csv +0 -0
  45. Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v1/experiments/grpo_lora_1000_20260407_192042/metrics_new.json +24 -0
  46. Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v1/experiments/grpo_lora_1000_20260407_192042/results_with_metrics_new.csv +0 -0
  47. Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v2/experiments/grpo_lora_1000_v2_20260407_202045/metrics_new.json +24 -0
  48. Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v2/experiments/grpo_lora_1000_v2_20260407_202045/results_with_metrics_new.csv +0 -0
  49. Llama3B/trainingExperiments/grpo_lora/EnglishEasy/350/v1/experiments/grpo_lora_350_20260403_174409/metrics_new.json +24 -0
  50. Llama3B/trainingExperiments/grpo_lora/EnglishEasy/350/v1/experiments/grpo_lora_350_20260403_174409/results_with_metrics_new.csv +0 -0
Llama3B/trainingExperiments/adapter/EnglishEasy/350/v2/experiments/adapter_20260404_190652/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.04700882641386074,
4
+ "recall": 1.0,
5
+ "f1": 0.08979642812539028
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.027316116377901274,
9
+ "recall": 0.5810848400556329,
10
+ "f1": 0.05217934307480954
11
+ },
12
+ "spriv": 0.0,
13
+ "domain_accuracy": 1.0,
14
+ "general_accuracy": 1.0,
15
+ "omr": 4.686227209791803,
16
+ "exact_copy_score": 0.007602434559002835,
17
+ "token_similarity": 0.048158546932332034,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 3595,
20
+ "total_pred_labels": 76475,
21
+ "clean_samples": 88,
22
+ "avg_latency_s": 7.934906908449613,
23
+ "total_time_s": 7141.416217604652
24
+ }
Llama3B/trainingExperiments/adapter/EnglishEasy/350/v2/experiments/adapter_20260404_190652/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_20260404_144425/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.05189983411011928,
4
+ "recall": 1.0,
5
+ "f1": 0.09867828176629617
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.033019985780867366,
9
+ "recall": 0.6362252663622526,
10
+ "f1": 0.0627816161009312
11
+ },
12
+ "spriv": 0.0,
13
+ "domain_accuracy": 1.0,
14
+ "general_accuracy": 1.0,
15
+ "omr": 3.9273300974387935,
16
+ "exact_copy_score": 0.009225880544613417,
17
+ "token_similarity": 0.038436360982926236,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 657,
20
+ "total_pred_labels": 12659,
21
+ "clean_samples": 16,
22
+ "avg_latency_s": 8.325605714871394,
23
+ "total_time_s": 1248.840857230709
24
+ }
Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_20260404_144425/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_test_20260420_124234/metrics_test_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.05375325597916173,
4
+ "recall": 1.0,
5
+ "f1": 0.10202247191011235
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.03496724287631226,
9
+ "recall": 0.6505139500734214,
10
+ "f1": 0.06636704119850188
11
+ },
12
+ "spriv": 0.0,
13
+ "domain_accuracy": 1.0,
14
+ "general_accuracy": 1.0,
15
+ "omr": 4.254654976198926,
16
+ "exact_copy_score": 0.008991691741189668,
17
+ "token_similarity": 0.04097306536992736,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 681,
20
+ "total_pred_labels": 12669,
21
+ "clean_samples": 7,
22
+ "avg_latency_s": 7.955297213972856,
23
+ "total_time_s": 1193.2945820959285
24
+ }
Llama3B/trainingExperiments/adapter/EnglishEasy/50/experiments/adapter_test_20260420_124234/results_test_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishEasy/50/v2/experiments/adapter_20260404_151236/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.0704001952171791,
4
+ "recall": 0.878234398782344,
5
+ "f1": 0.13035129334688805
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.044045876037091265,
9
+ "recall": 0.5494672754946728,
10
+ "f1": 0.0815542753868745
11
+ },
12
+ "spriv": 0.121765601217656,
13
+ "domain_accuracy": 0.9037656903765691,
14
+ "general_accuracy": 0.8636363636363636,
15
+ "omr": 2.0283767862572213,
16
+ "exact_copy_score": 0.008728569217721697,
17
+ "token_similarity": 0.00677797629799245,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 657,
20
+ "total_pred_labels": 8196,
21
+ "clean_samples": 16,
22
+ "avg_latency_s": 8.44037366228023,
23
+ "total_time_s": 1266.0560493420344
24
+ }
Llama3B/trainingExperiments/adapter/EnglishEasy/50/v2/experiments/adapter_20260404_151236/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/1000/experiments/adapter_20260405_211432/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.09571663291429444,
4
+ "recall": 0.19491129785247432,
5
+ "f1": 0.12838582374497093
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.05502273508845669,
9
+ "recall": 0.11204481792717087,
10
+ "f1": 0.07380262921866591
11
+ },
12
+ "spriv": 0.8050887021475257,
13
+ "domain_accuracy": 0.136431784107946,
14
+ "general_accuracy": 0.22827813760234633,
15
+ "omr": 1.634683720186068,
16
+ "exact_copy_score": 0.003227267730303486,
17
+ "token_similarity": 0.005893086323646291,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 12852,
20
+ "total_pred_labels": 26171,
21
+ "clean_samples": 302,
22
+ "avg_latency_s": 1.2711757163944033,
23
+ "total_time_s": 3813.52714918321
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/1000/experiments/adapter_20260405_211432/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_20260406_013129/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.07211714014459943,
4
+ "recall": 0.9220354808590103,
5
+ "f1": 0.13377133568138716
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.02959541371503688,
9
+ "recall": 0.3783846872082166,
10
+ "f1": 0.054897046870766734
11
+ },
12
+ "spriv": 0.07796451914098973,
13
+ "domain_accuracy": 0.9246091240094239,
14
+ "general_accuracy": 0.9205670292068924,
15
+ "omr": 3.193156140114491,
16
+ "exact_copy_score": 0.009198880371530127,
17
+ "token_similarity": 0.031767798836708626,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 12852,
20
+ "total_pred_labels": 164316,
21
+ "clean_samples": 302,
22
+ "avg_latency_s": 7.990792127518216,
23
+ "total_time_s": 23972.37638255465
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_20260406_013129/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_test_20260420_235923/metrics_test_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.07168801262952984,
4
+ "recall": 0.933029507183686,
5
+ "f1": 0.13314594356261023
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.029134568589979464,
9
+ "recall": 0.37919048354704155,
10
+ "f1": 0.05411155202821869
11
+ },
12
+ "spriv": 0.06697049281631393,
13
+ "domain_accuracy": 0.9344741235392321,
14
+ "general_accuracy": 0.9321805248957566,
15
+ "omr": 3.1224575991531607,
16
+ "exact_copy_score": 0.00951542718072523,
17
+ "token_similarity": 0.03164036917012624,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 12946,
20
+ "total_pred_labels": 168494,
21
+ "clean_samples": 300,
22
+ "avg_latency_s": 8.077805570261242,
23
+ "total_time_s": 24233.416710783727
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/1000/v2/experiments/adapter_test_20260420_235923/results_test_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/350/experiments/adapter_20260405_133314/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.04427059699054326,
4
+ "recall": 0.8381443298969072,
5
+ "f1": 0.08409909832250057
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.0,
9
+ "recall": 0.0,
10
+ "f1": 0.0
11
+ },
12
+ "spriv": 0.1618556701030928,
13
+ "domain_accuracy": 0.8072122052704577,
14
+ "general_accuracy": 0.8685286103542235,
15
+ "omr": 3.5983006851491774,
16
+ "exact_copy_score": 0.007929221019225987,
17
+ "token_similarity": 0.014194254677781301,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 2910,
20
+ "total_pred_labels": 55093,
21
+ "clean_samples": 103,
22
+ "avg_latency_s": 6.803836256845906,
23
+ "total_time_s": 6123.452631161315
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/350/experiments/adapter_20260405_133314/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_20260405_155859/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.03941113326176967,
4
+ "recall": 0.9714776632302405,
5
+ "f1": 0.07574925309146448
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.002105087061382108,
9
+ "recall": 0.05189003436426117,
10
+ "f1": 0.004046033681220777
11
+ },
12
+ "spriv": 0.02852233676975945,
13
+ "domain_accuracy": 0.9680998613037448,
14
+ "general_accuracy": 0.9747956403269755,
15
+ "omr": 4.261118945495542,
16
+ "exact_copy_score": 0.007033823044061526,
17
+ "token_similarity": 0.002226409081173069,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 2910,
20
+ "total_pred_labels": 71731,
21
+ "clean_samples": 103,
22
+ "avg_latency_s": 8.267505045320787,
23
+ "total_time_s": 7440.754540788708
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_20260405_155859/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_test_20260420_215936/metrics_test_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.04060906672578142,
4
+ "recall": 0.9702085402184707,
5
+ "f1": 0.07795523757596712
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.0022168033695411215,
9
+ "recall": 0.052962595167163194,
10
+ "f1": 0.004255488915781215
11
+ },
12
+ "spriv": 0.029791459781529295,
13
+ "domain_accuracy": 0.9637203166226913,
14
+ "general_accuracy": 0.9767441860465116,
15
+ "omr": 4.307282408481972,
16
+ "exact_copy_score": 0.007229752497285992,
17
+ "token_similarity": 0.0019987081407696564,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 3021,
20
+ "total_pred_labels": 72176,
21
+ "clean_samples": 79,
22
+ "avg_latency_s": 7.9535390527484315,
23
+ "total_time_s": 7158.185147473589
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/350/v2/experiments/adapter_test_20260420_215936/results_test_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/50/experiments/adapter_20260405_120144/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.048,
4
+ "recall": 1.0,
5
+ "f1": 0.0916030534351145
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.02831372549019608,
9
+ "recall": 0.5898692810457516,
10
+ "f1": 0.05403382727136657
11
+ },
12
+ "spriv": 0.0,
13
+ "domain_accuracy": 1.0,
14
+ "general_accuracy": 1.0,
15
+ "omr": 4.455639572415888,
16
+ "exact_copy_score": 0.007472660488614406,
17
+ "token_similarity": 0.0462626832624723,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 612,
20
+ "total_pred_labels": 12750,
21
+ "clean_samples": 14,
22
+ "avg_latency_s": 8.343479785760865,
23
+ "total_time_s": 1251.5219678641297
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/50/experiments/adapter_20260405_120144/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_20260405_123002/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.06487861761899713,
4
+ "recall": 1.0,
5
+ "f1": 0.1218516674962668
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.038269903530160077,
9
+ "recall": 0.5898692810457516,
10
+ "f1": 0.07187655550024889
11
+ },
12
+ "spriv": 0.0,
13
+ "domain_accuracy": 1.0,
14
+ "general_accuracy": 1.0,
15
+ "omr": 3.689442551342706,
16
+ "exact_copy_score": 0.00903182681087819,
17
+ "token_similarity": 0.07343849911424274,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 612,
20
+ "total_pred_labels": 9433,
21
+ "clean_samples": 14,
22
+ "avg_latency_s": 7.949676980772056,
23
+ "total_time_s": 1192.4515471158084
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_20260405_123002/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_test_20260420_213827/metrics_test_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.06503288775726714,
4
+ "recall": 0.9967479674796748,
5
+ "f1": 0.12209939249078777
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.039253129641417354,
9
+ "recall": 0.6016260162601627,
10
+ "f1": 0.07369783886067124
11
+ },
12
+ "spriv": 0.0032520325203252032,
13
+ "domain_accuracy": 0.9918367346938776,
14
+ "general_accuracy": 1.0,
15
+ "omr": 3.748759679310321,
16
+ "exact_copy_score": 0.008714729621792408,
17
+ "token_similarity": 0.07601194714460285,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 615,
20
+ "total_pred_labels": 9426,
21
+ "clean_samples": 14,
22
+ "avg_latency_s": 8.299127354286611,
23
+ "total_time_s": 1244.8691031429917
24
+ }
Llama3B/trainingExperiments/adapter/EnglishHard/50/v2/experiments/adapter_test_20260420_213827/results_test_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_20260408_144903/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.06896551724137931,
4
+ "recall": 0.8997481901164621,
5
+ "f1": 0.12811132710741116
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.005307823611370806,
9
+ "recall": 0.06924771797293043,
10
+ "f1": 0.009859888740119104
11
+ },
12
+ "spriv": 0.10025180988353793,
13
+ "domain_accuracy": 0.898909202642495,
14
+ "general_accuracy": 0.900629133731247,
15
+ "omr": 2.127333578718273,
16
+ "exact_copy_score": 0.0034113359955505527,
17
+ "token_similarity": 0.016394265162782296,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 12708,
20
+ "total_pred_labels": 165793,
21
+ "clean_samples": 72,
22
+ "avg_latency_s": 7.390383391882448,
23
+ "total_time_s": 22171.150175647344
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_20260408_144903/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_test_20260421_090302/metrics_test_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.06884549511666595,
4
+ "recall": 0.8850919422302896,
5
+ "f1": 0.127753855967922
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.0059115659204056455,
9
+ "recall": 0.07600031568147739,
10
+ "f1": 0.01096985851958171
11
+ },
12
+ "spriv": 0.11490805776971036,
13
+ "domain_accuracy": 0.8894448788115715,
14
+ "general_accuracy": 0.8806564690885914,
15
+ "omr": 2.0682823583141436,
16
+ "exact_copy_score": 0.0033614395055335733,
17
+ "token_similarity": 0.018327876195571195,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 12671,
20
+ "total_pred_labels": 162901,
21
+ "clean_samples": 77,
22
+ "avg_latency_s": 7.335871835032168,
23
+ "total_time_s": 22007.615505096503
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/1000/experiments/adapter_test_20260421_090302/results_test_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/1000/v2/experiments/adapter_20260409_001233/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.06166285019647754,
4
+ "recall": 0.8581995593327038,
5
+ "f1": 0.11505857901705412
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.010239448166680801,
9
+ "recall": 0.14250865596474663,
10
+ "f1": 0.019106096332283606
11
+ },
12
+ "spriv": 0.1418004406672962,
13
+ "domain_accuracy": 0.8417575664464587,
14
+ "general_accuracy": 0.875463784481368,
15
+ "omr": 2.3693660228016773,
16
+ "exact_copy_score": 0.004346280492410998,
17
+ "token_similarity": 0.01699594990089344,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 12708,
20
+ "total_pred_labels": 176865,
21
+ "clean_samples": 72,
22
+ "avg_latency_s": 8.018695728058306,
23
+ "total_time_s": 24056.087184174918
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/1000/v2/experiments/adapter_20260409_001233/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/350/experiments/adapter_20260408_065136/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.048752849582061296,
4
+ "recall": 0.9886455798864558,
5
+ "f1": 0.0929233896582391
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.017624081801335803,
9
+ "recall": 0.3573938902406056,
10
+ "f1": 0.03359166560792784
11
+ },
12
+ "spriv": 0.0113544201135442,
13
+ "domain_accuracy": 0.9878849066128218,
14
+ "general_accuracy": 0.989522700814901,
15
+ "omr": 4.366632793961803,
16
+ "exact_copy_score": 0.004074741201694229,
17
+ "token_similarity": 0.02748890864775017,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 3699,
20
+ "total_pred_labels": 75011,
21
+ "clean_samples": 8,
22
+ "avg_latency_s": 8.117203779817403,
23
+ "total_time_s": 7305.483401835663
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/350/experiments/adapter_20260408_065136/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_20260408_093702/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.065063530400084,
4
+ "recall": 0.8375236550419032,
5
+ "f1": 0.12074677475932495
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.020854772655675733,
9
+ "recall": 0.2684509326845093,
10
+ "f1": 0.038702888100713255
11
+ },
12
+ "spriv": 0.1624763449580968,
13
+ "domain_accuracy": 0.8404846037354872,
14
+ "general_accuracy": 0.8341094295692666,
15
+ "omr": 4.077062563353836,
16
+ "exact_copy_score": 0.006450372269713989,
17
+ "token_similarity": 0.02996456406904448,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 3699,
20
+ "total_pred_labels": 47615,
21
+ "clean_samples": 8,
22
+ "avg_latency_s": 8.070445773015058,
23
+ "total_time_s": 7263.401195713552
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_20260408_093702/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_test_20260421_070514/metrics_test_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.06477030793880731,
4
+ "recall": 0.8032117583015786,
5
+ "f1": 0.1198740733218239
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.02100480674260881,
9
+ "recall": 0.26047904191616766,
10
+ "f1": 0.03887478419823296
11
+ },
12
+ "spriv": 0.19678824169842135,
13
+ "domain_accuracy": 0.80370942812983,
14
+ "general_accuracy": 0.8026543566070398,
15
+ "omr": 4.479607320354225,
16
+ "exact_copy_score": 0.006193275303346803,
17
+ "token_similarity": 0.03037843051882257,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 3674,
20
+ "total_pred_labels": 45561,
21
+ "clean_samples": 20,
22
+ "avg_latency_s": 7.82180106826851,
23
+ "total_time_s": 7039.620961441658
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/350/v2/experiments/adapter_test_20260421_070514/results_test_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_20260408_051945/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.05883462191212521,
4
+ "recall": 1.0,
5
+ "f1": 0.11113089937666963
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.02187441071091835,
9
+ "recall": 0.3717948717948718,
10
+ "f1": 0.041317898486197685
11
+ },
12
+ "spriv": 0.0,
13
+ "domain_accuracy": 1.0,
14
+ "general_accuracy": 1.0,
15
+ "omr": 0.0,
16
+ "exact_copy_score": 0.007504476263131148,
17
+ "token_similarity": 0.04525784133237164,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 624,
20
+ "total_pred_labels": 10606,
21
+ "clean_samples": 0,
22
+ "avg_latency_s": 8.501927905186701,
23
+ "total_time_s": 1275.289185778005
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_20260408_051945/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_test_20260421_064404/metrics_test_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.057937319817725284,
4
+ "recall": 1.0,
5
+ "f1": 0.10952883263009845
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.023807309588021946,
9
+ "recall": 0.41091492776886035,
10
+ "f1": 0.045007032348804495
11
+ },
12
+ "spriv": 0.0,
13
+ "domain_accuracy": 1.0,
14
+ "general_accuracy": 1.0,
15
+ "omr": 0.0,
16
+ "exact_copy_score": 0.00683287850455636,
17
+ "token_similarity": 0.04850282318059296,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 623,
20
+ "total_pred_labels": 10753,
21
+ "clean_samples": 0,
22
+ "avg_latency_s": 8.319182492991288,
23
+ "total_time_s": 1247.8773739486933
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/50/experiments/adapter_test_20260421_064404/results_test_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/adapter/Mixed/50/v2/experiments/adapter_20260408_054820/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 1.0,
4
+ "recall": 0.21955128205128205,
5
+ "f1": 0.36005256241787126
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.3284671532846715,
9
+ "recall": 0.07211538461538461,
10
+ "f1": 0.11826544021024965
11
+ },
12
+ "spriv": 0.780448717948718,
13
+ "domain_accuracy": 0.2713864306784661,
14
+ "general_accuracy": 0.15789473684210525,
15
+ "omr": 0.0,
16
+ "exact_copy_score": 0.012294157436718505,
17
+ "token_similarity": 0.0017016806722689075,
18
+ "n_samples": 150,
19
+ "total_gt_labels": 624,
20
+ "total_pred_labels": 137,
21
+ "clean_samples": 0,
22
+ "avg_latency_s": 7.993254954383398,
23
+ "total_time_s": 1198.9882431575097
24
+ }
Llama3B/trainingExperiments/adapter/Mixed/50/v2/experiments/adapter_20260408_054820/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v1/experiments/grpo_lora_1000_20260407_192042/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.9253134457414613,
4
+ "recall": 0.8219875180028805,
5
+ "f1": 0.8705954136370572
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.5420449632511889,
9
+ "recall": 0.4815170427268363,
10
+ "f1": 0.5099913560787105
11
+ },
12
+ "spriv": 0.17801248199711953,
13
+ "domain_accuracy": 0.6299262381454163,
14
+ "general_accuracy": 0.9827160493827161,
15
+ "omr": 0.0032706261465018428,
16
+ "exact_copy_score": 0.20174788148463157,
17
+ "token_similarity": 0.666371538709397,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 10415,
20
+ "total_pred_labels": 9252,
21
+ "clean_samples": 293,
22
+ "avg_latency_s": 1.193282431629603,
23
+ "total_time_s": 3579.8472948888084
24
+ }
Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v1/experiments/grpo_lora_1000_20260407_192042/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v2/experiments/grpo_lora_1000_v2_20260407_202045/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.9286075437354636,
4
+ "recall": 0.8817090734517523,
5
+ "f1": 0.9045508274231678
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.5148144402871878,
9
+ "recall": 0.4888142102736438,
10
+ "f1": 0.5014775413711584
11
+ },
12
+ "spriv": 0.11829092654824772,
13
+ "domain_accuracy": 0.7494204425711275,
14
+ "general_accuracy": 0.9924162257495591,
15
+ "omr": 0.0003792188092529389,
16
+ "exact_copy_score": 0.21840731310487915,
17
+ "token_similarity": 0.6880078578509489,
18
+ "n_samples": 3000,
19
+ "total_gt_labels": 10415,
20
+ "total_pred_labels": 9889,
21
+ "clean_samples": 293,
22
+ "avg_latency_s": 1.232007218956055,
23
+ "total_time_s": 3696.0216568681644
24
+ }
Llama3B/trainingExperiments/grpo_lora/EnglishEasy/1000/v2/experiments/grpo_lora_1000_v2_20260407_202045/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff
 
Llama3B/trainingExperiments/grpo_lora/EnglishEasy/350/v1/experiments/grpo_lora_350_20260403_174409/metrics_new.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "span_detection": {
3
+ "precision": 0.939094387755102,
4
+ "recall": 0.8191933240611962,
5
+ "f1": 0.8750557123755758
6
+ },
7
+ "label_accuracy": {
8
+ "precision": 0.5105229591836735,
9
+ "recall": 0.44534075104311543,
10
+ "f1": 0.47570940424899716
11
+ },
12
+ "spriv": 0.1808066759388039,
13
+ "domain_accuracy": 0.7788911155644622,
14
+ "general_accuracy": 0.847950428979981,
15
+ "omr": 0.003082826577056805,
16
+ "exact_copy_score": 0.20257382241578736,
17
+ "token_similarity": 0.6256849928495761,
18
+ "n_samples": 900,
19
+ "total_gt_labels": 3595,
20
+ "total_pred_labels": 3136,
21
+ "clean_samples": 88,
22
+ "avg_latency_s": 1.2407716992668187,
23
+ "total_time_s": 1116.6945293401368
24
+ }
Llama3B/trainingExperiments/grpo_lora/EnglishEasy/350/v1/experiments/grpo_lora_350_20260403_174409/results_with_metrics_new.csv ADDED
The diff for this file is too large to render. See raw diff