Buckets:

Mercity
/

policy-classifier

Files
xet
Mercity/policy-classifier / 03-11-progress /modernbert.log
Rishik001
5 days ago
download
raw
78 kB
{
  "train": {
    "pairs": 27000,
    "labels": {
      "followed": 10087,
      "not_applicable": 10086,
      "violated": 6827
    }
  },
  "validation": {
    "pairs": 3000,
    "labels": {
      "not_applicable": 1115,
      "violated": 770,
      "followed": 1115
    }
  },
  "test_ood": {
    "pairs": 0,
    "labels": {}
  }
}
`torch_dtype` is deprecated! Use `dtype` instead!
Some weights of ModernBertForSequenceClassification were not initialized from the model checkpoint at answerdotai/ModernBERT-large and are newly initialized: ['classifier.bias', 'classifier.weight']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
trainable params: 15,446,019 || all params: 411,280,390 || trainable%: 3.7556
⚙️  Running in WANDB offline mode

Map:   0%|          | 0/27000 [00:00<?, ? examples/s]
Map:   4%|▎         | 1000/27000 [00:00<00:15, 1657.77 examples/s]
Map:   7%|▋         | 2000/27000 [00:01<00:17, 1469.35 examples/s]
Map:  11%|█         | 3000/27000 [00:01<00:13, 1801.10 examples/s]
Map:  15%|█▍        | 4000/27000 [00:02<00:10, 2123.08 examples/s]
Map:  19%|█▊        | 5000/27000 [00:02<00:08, 2568.82 examples/s]
Map:  22%|██▏       | 6000/27000 [00:02<00:08, 2574.23 examples/s]
Map:  26%|██▌       | 7000/27000 [00:03<00:09, 2078.83 examples/s]
Map:  30%|██▉       | 8000/27000 [00:03<00:08, 2302.47 examples/s]
Map:  33%|███▎      | 9000/27000 [00:04<00:07, 2506.00 examples/s]
Map:  37%|███▋      | 10000/27000 [00:04<00:06, 2723.28 examples/s]
Map:  41%|████      | 11000/27000 [00:04<00:05, 3110.38 examples/s]
Map:  44%|████▍     | 12000/27000 [00:04<00:05, 2867.29 examples/s]
Map:  48%|████▊     | 13000/27000 [00:05<00:07, 1818.10 examples/s]
Map:  52%|█████▏    | 14000/27000 [00:06<00:06, 2112.87 examples/s]
Map:  56%|█████▌    | 15000/27000 [00:06<00:05, 2356.81 examples/s]
Map:  59%|█████▉    | 16000/27000 [00:06<00:04, 2601.17 examples/s]
Map:  63%|██████▎   | 17000/27000 [00:07<00:03, 3001.32 examples/s]
Map:  67%|██████▋   | 18000/27000 [00:07<00:02, 3095.93 examples/s]
Map:  70%|███████   | 19000/27000 [00:07<00:02, 3198.63 examples/s]
Map:  74%|███████▍  | 20000/27000 [00:07<00:02, 3394.63 examples/s]
Map:  78%|███████▊  | 21000/27000 [00:08<00:01, 3614.13 examples/s]
Map:  81%|████████▏ | 22000/27000 [00:08<00:01, 2880.17 examples/s]
Map:  85%|████████▌ | 23000/27000 [00:08<00:01, 3201.44 examples/s]
Map:  89%|████████▉ | 24000/27000 [00:09<00:00, 3501.02 examples/s]
Map:  93%|█████████▎| 25000/27000 [00:09<00:00, 3992.64 examples/s]
Map:  96%|█████████▋| 26000/27000 [00:10<00:00, 2063.28 examples/s]
Map: 100%|██████████| 27000/27000 [00:10<00:00, 2560.50 examples/s]
Map: 100%|██████████| 27000/27000 [00:10<00:00, 2565.74 examples/s]

Map:   0%|          | 0/3000 [00:00<?, ? examples/s]
Map:  33%|███▎      | 1000/3000 [00:00<00:00, 4144.00 examples/s]
Map:  67%|██████▋   | 2000/3000 [00:00<00:00, 4076.66 examples/s]
Map: 100%|██████████| 3000/3000 [00:00<00:00, 4525.08 examples/s]
Map: 100%|██████████| 3000/3000 [00:00<00:00, 4356.53 examples/s]
wandb: Tracking run with wandb version 0.31.0
wandb: W&B syncing is set to `offline` in this directory. Run `wandb online` or set WANDB_MODE=online to enable cloud syncing.
wandb: Run data is saved locally in /workspace/wandb/offline-run-20261103_143023-clywa1fj
wandb: View this run in the terminal with `wandb leet`
wandb: WARNING URL not available in offline run
The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'eos_token_id': None, 'bos_token_id': None}.

  0%|          | 0/633 [00:00<?, ?it/s]
  0%|          | 1/633 [00:04<46:58,  4.46s/it]
  0%|          | 2/633 [00:05<28:08,  2.68s/it]
  0%|          | 3/633 [00:06<20:36,  1.96s/it]
  1%|          | 4/633 [00:08<16:58,  1.62s/it]
  1%|          | 5/633 [00:09<15:58,  1.53s/it]
  1%|          | 6/633 [00:11<16:06,  1.54s/it]
  1%|          | 7/633 [00:11<13:42,  1.31s/it]
  1%|▏         | 8/633 [00:13<14:10,  1.36s/it]
  1%|▏         | 9/633 [00:14<14:00,  1.35s/it]
  2%|▏         | 10/633 [00:15<12:54,  1.24s/it]
                                                
{'loss': 1.1116, 'grad_norm': 1.9982777833938599, 'learning_rate': 5.6250000000000005e-05, 'epoch': 0.02}

  2%|▏         | 10/633 [00:15<12:54,  1.24s/it]
  2%|▏         | 11/633 [00:16<12:44,  1.23s/it]
  2%|▏         | 12/633 [00:18<13:41,  1.32s/it]
  2%|▏         | 13/633 [00:19<13:24,  1.30s/it]
  2%|▏         | 14/633 [00:20<12:59,  1.26s/it]
  2%|▏         | 15/633 [00:22<12:53,  1.25s/it]
  3%|▎         | 16/633 [00:23<14:31,  1.41s/it]
  3%|▎         | 17/633 [00:25<16:36,  1.62s/it]
  3%|▎         | 18/633 [00:27<16:45,  1.63s/it]
  3%|▎         | 19/633 [00:28<15:29,  1.51s/it]
  3%|▎         | 20/633 [00:30<16:34,  1.62s/it]
                                                
{'loss': 1.0984, 'grad_norm': 4.667947292327881, 'learning_rate': 0.00011875, 'epoch': 0.05}

  3%|▎         | 20/633 [00:30<16:34,  1.62s/it]
  3%|▎         | 21/633 [00:32<16:29,  1.62s/it]
  3%|▎         | 22/633 [00:33<14:12,  1.40s/it]
  4%|▎         | 23/633 [00:34<14:59,  1.47s/it]
  4%|▍         | 24/633 [00:35<13:08,  1.30s/it]
  4%|▍         | 25/633 [00:36<12:11,  1.20s/it]
  4%|▍         | 26/633 [00:37<12:07,  1.20s/it]
  4%|▍         | 27/633 [00:39<12:42,  1.26s/it]
  4%|▍         | 28/633 [00:40<12:55,  1.28s/it]
  5%|▍         | 29/633 [00:41<12:14,  1.22s/it]
  5%|▍         | 30/633 [00:42<11:07,  1.11s/it]
                                                
{'loss': 1.0464, 'grad_norm': 3.6819779872894287, 'learning_rate': 0.00018125000000000001, 'epoch': 0.07}

  5%|▍         | 30/633 [00:42<11:07,  1.11s/it]
  5%|▍         | 31/633 [00:44<12:17,  1.23s/it]
  5%|▌         | 32/633 [00:45<12:01,  1.20s/it]
  5%|▌         | 33/633 [00:46<12:26,  1.24s/it]
  5%|▌         | 34/633 [00:47<12:20,  1.24s/it]
  6%|▌         | 35/633 [00:48<12:00,  1.21s/it]
  6%|▌         | 36/633 [00:50<12:01,  1.21s/it]
  6%|▌         | 37/633 [00:51<11:09,  1.12s/it]
  6%|▌         | 38/633 [00:52<11:51,  1.20s/it]
  6%|▌         | 39/633 [00:54<13:54,  1.41s/it]
  6%|▋         | 40/633 [00:55<12:11,  1.23s/it]
                                                
{'loss': 0.9683, 'grad_norm': 3.145085334777832, 'learning_rate': 0.00019767054908485858, 'epoch': 0.09}

  6%|▋         | 40/633 [00:55<12:11,  1.23s/it]
  6%|▋         | 41/633 [00:56<13:45,  1.39s/it]
  7%|▋         | 42/633 [00:58<15:23,  1.56s/it]
  7%|▋         | 43/633 [01:00<15:21,  1.56s/it]
  7%|▋         | 44/633 [01:02<16:37,  1.69s/it]
  7%|▋         | 45/633 [01:03<15:09,  1.55s/it]
  7%|▋         | 46/633 [01:05<16:26,  1.68s/it]
  7%|▋         | 47/633 [01:06<14:26,  1.48s/it]
  8%|▊         | 48/633 [01:08<15:01,  1.54s/it]
  8%|▊         | 49/633 [01:09<13:27,  1.38s/it]
  8%|▊         | 50/633 [01:10<11:48,  1.22s/it]
                                                
{'loss': 0.8828, 'grad_norm': 2.9893064498901367, 'learning_rate': 0.00019434276206322797, 'epoch': 0.12}

  8%|▊         | 50/633 [01:10<11:48,  1.22s/it]
  8%|▊         | 51/633 [01:11<11:45,  1.21s/it]
  8%|▊         | 52/633 [01:12<11:44,  1.21s/it]
  8%|▊         | 53/633 [01:13<11:30,  1.19s/it]
  9%|▊         | 54/633 [01:15<11:57,  1.24s/it]
  9%|▊         | 55/633 [01:16<11:31,  1.20s/it]
  9%|▉         | 56/633 [01:17<11:58,  1.25s/it]
  9%|▉         | 57/633 [01:19<13:14,  1.38s/it]
  9%|▉         | 58/633 [01:20<12:52,  1.34s/it]
  9%|▉         | 59/633 [01:22<13:42,  1.43s/it]
  9%|▉         | 60/633 [01:23<12:31,  1.31s/it]
                                                
{'loss': 0.8175, 'grad_norm': 3.7744088172912598, 'learning_rate': 0.00019101497504159736, 'epoch': 0.14}

  9%|▉         | 60/633 [01:23<12:31,  1.31s/it]
 10%|▉         | 61/633 [01:24<12:50,  1.35s/it]
 10%|▉         | 62/633 [01:25<11:43,  1.23s/it]
 10%|▉         | 63/633 [01:26<11:17,  1.19s/it]
 10%|█         | 64/633 [01:28<11:55,  1.26s/it]
 10%|█         | 65/633 [01:29<13:22,  1.41s/it]
 10%|█         | 66/633 [01:31<13:35,  1.44s/it]
 11%|█         | 67/633 [01:33<14:20,  1.52s/it]
 11%|█         | 68/633 [01:33<12:30,  1.33s/it]
 11%|█         | 69/633 [01:35<12:07,  1.29s/it]
 11%|█         | 70/633 [01:36<11:05,  1.18s/it]
                                                
{'loss': 0.7834, 'grad_norm': 2.660766363143921, 'learning_rate': 0.00018768718801996675, 'epoch': 0.17}

 11%|█         | 70/633 [01:36<11:05,  1.18s/it]
 11%|█         | 71/633 [01:37<11:21,  1.21s/it]
 11%|█▏        | 72/633 [01:38<11:33,  1.24s/it]
 12%|█▏        | 73/633 [01:40<12:10,  1.30s/it]
 12%|█▏        | 74/633 [01:41<12:04,  1.30s/it]
 12%|█▏        | 75/633 [01:42<10:52,  1.17s/it]
 12%|█▏        | 76/633 [01:43<12:18,  1.33s/it]
 12%|█▏        | 77/633 [01:45<12:27,  1.34s/it]
 12%|█▏        | 78/633 [01:46<12:30,  1.35s/it]
 12%|█▏        | 79/633 [01:48<12:59,  1.41s/it]
 13%|█▎        | 80/633 [01:49<11:27,  1.24s/it]
                                                
{'loss': 0.7211, 'grad_norm': 1.944366693496704, 'learning_rate': 0.00018435940099833611, 'epoch': 0.19}

 13%|█▎        | 80/633 [01:49<11:27,  1.24s/it]
 13%|█▎        | 81/633 [01:49<10:24,  1.13s/it]
 13%|█▎        | 82/633 [01:51<10:22,  1.13s/it]
 13%|█▎        | 83/633 [01:52<12:36,  1.37s/it]
 13%|█▎        | 84/633 [01:54<12:13,  1.34s/it]
 13%|█▎        | 85/633 [01:55<12:29,  1.37s/it]
 14%|█▎        | 86/633 [01:56<10:49,  1.19s/it]
 14%|█▎        | 87/633 [01:58<12:04,  1.33s/it]
 14%|█▍        | 88/633 [01:59<13:03,  1.44s/it]
 14%|█▍        | 89/633 [02:01<13:13,  1.46s/it]
 14%|█▍        | 90/633 [02:02<11:31,  1.27s/it]
                                                
{'loss': 0.6444, 'grad_norm': 20.749786376953125, 'learning_rate': 0.0001810316139767055, 'epoch': 0.21}

 14%|█▍        | 90/633 [02:02<11:31,  1.27s/it]
 14%|█▍        | 91/633 [02:03<10:47,  1.19s/it]
 15%|█▍        | 92/633 [02:04<11:02,  1.23s/it]
 15%|█▍        | 93/633 [02:05<10:02,  1.12s/it]
 15%|█▍        | 94/633 [02:06<09:49,  1.09s/it]
 15%|█▌        | 95/633 [02:07<10:16,  1.15s/it]
 15%|█▌        | 96/633 [02:09<11:57,  1.34s/it]
 15%|█▌        | 97/633 [02:10<11:41,  1.31s/it]
 15%|█▌        | 98/633 [02:12<12:04,  1.35s/it]
 16%|█▌        | 99/633 [02:13<12:32,  1.41s/it]
 16%|█▌        | 100/633 [02:14<12:22,  1.39s/it]
                                                 
{'loss': 0.6188, 'grad_norm': 1.806443691253662, 'learning_rate': 0.00017770382695507487, 'epoch': 0.24}

 16%|█▌        | 100/633 [02:15<12:22,  1.39s/it]
 16%|█▌        | 101/633 [02:16<11:40,  1.32s/it]
 16%|█▌        | 102/633 [02:17<12:05,  1.37s/it]
 16%|█▋        | 103/633 [02:19<12:52,  1.46s/it]
 16%|█▋        | 104/633 [02:20<12:40,  1.44s/it]
 17%|█▋        | 105/633 [02:22<13:42,  1.56s/it]
 17%|█▋        | 106/633 [02:23<12:55,  1.47s/it]
 17%|█▋        | 107/633 [02:25<14:10,  1.62s/it]
 17%|█▋        | 108/633 [02:27<13:12,  1.51s/it]
 17%|█▋        | 109/633 [02:28<12:44,  1.46s/it]
 17%|█▋        | 110/633 [02:30<13:57,  1.60s/it]
                                                 
{'loss': 0.6589, 'grad_norm': 3.5172643661499023, 'learning_rate': 0.00017437603993344426, 'epoch': 0.26}

 17%|█▋        | 110/633 [02:30<13:57,  1.60s/it]
 18%|█▊        | 111/633 [02:33<17:03,  1.96s/it]
 18%|█▊        | 112/633 [02:34<15:42,  1.81s/it]
 18%|█▊        | 113/633 [02:35<13:49,  1.60s/it]
 18%|█▊        | 114/633 [02:37<13:26,  1.55s/it]
 18%|█▊        | 115/633 [02:38<14:17,  1.66s/it]
 18%|█▊        | 116/633 [02:40<13:21,  1.55s/it]
 18%|█▊        | 117/633 [02:41<11:28,  1.33s/it]
 19%|█▊        | 118/633 [02:42<10:48,  1.26s/it]
 19%|█▉        | 119/633 [02:43<09:53,  1.15s/it]
 19%|█▉        | 120/633 [02:44<09:42,  1.14s/it]
                                                 
{'loss': 0.5931, 'grad_norm': 1.5550144910812378, 'learning_rate': 0.00017104825291181365, 'epoch': 0.28}

 19%|█▉        | 120/633 [02:44<09:42,  1.14s/it]
 19%|█▉        | 121/633 [02:45<09:55,  1.16s/it]
 19%|█▉        | 122/633 [02:46<09:53,  1.16s/it]
 19%|█▉        | 123/633 [02:47<09:59,  1.18s/it]
 20%|█▉        | 124/633 [02:49<10:56,  1.29s/it]
 20%|█▉        | 125/633 [02:50<09:59,  1.18s/it]
 20%|█▉        | 126/633 [02:51<10:37,  1.26s/it]
 20%|██        | 127/633 [02:52<10:33,  1.25s/it]
 20%|██        | 128/633 [02:54<10:02,  1.19s/it]
 20%|██        | 129/633 [02:55<09:44,  1.16s/it]
 21%|██        | 130/633 [02:56<09:56,  1.19s/it]
                                                 
{'loss': 0.5718, 'grad_norm': 4.285693168640137, 'learning_rate': 0.00016772046589018304, 'epoch': 0.31}

 21%|██        | 130/633 [02:56<09:56,  1.19s/it]
 21%|██        | 131/633 [02:57<10:05,  1.21s/it]
 21%|██        | 132/633 [03:00<14:21,  1.72s/it]
 21%|██        | 133/633 [03:02<14:30,  1.74s/it]
 21%|██        | 134/633 [03:03<13:57,  1.68s/it]
 21%|██▏       | 135/633 [03:04<12:30,  1.51s/it]
 21%|██▏       | 136/633 [03:05<10:38,  1.28s/it]
 22%|██▏       | 137/633 [03:06<10:21,  1.25s/it]
 22%|██▏       | 138/633 [03:08<10:57,  1.33s/it]
 22%|██▏       | 139/633 [03:09<10:56,  1.33s/it]
 22%|██▏       | 140/633 [03:10<10:46,  1.31s/it]
                                                 
{'loss': 0.5715, 'grad_norm': 2.0089755058288574, 'learning_rate': 0.00016439267886855243, 'epoch': 0.33}

 22%|██▏       | 140/633 [03:10<10:46,  1.31s/it]
 22%|██▏       | 141/633 [03:12<10:50,  1.32s/it]
 22%|██▏       | 142/633 [03:13<11:01,  1.35s/it]
 23%|██▎       | 143/633 [03:14<10:33,  1.29s/it]
 23%|██▎       | 144/633 [03:16<11:30,  1.41s/it]
 23%|██▎       | 145/633 [03:17<11:04,  1.36s/it]
 23%|██▎       | 146/633 [03:19<10:39,  1.31s/it]
 23%|██▎       | 147/633 [03:20<10:48,  1.34s/it]
 23%|██▎       | 148/633 [03:22<11:36,  1.44s/it]
 24%|██▎       | 149/633 [03:24<12:55,  1.60s/it]
 24%|██▎       | 150/633 [03:25<11:48,  1.47s/it]
                                                 
{'loss': 0.5804, 'grad_norm': 2.3266260623931885, 'learning_rate': 0.0001610648918469218, 'epoch': 0.36}

 24%|██▎       | 150/633 [03:25<11:48,  1.47s/it]
 24%|██▍       | 151/633 [03:26<11:37,  1.45s/it]
 24%|██▍       | 152/633 [03:27<10:28,  1.31s/it]
 24%|██▍       | 153/633 [03:28<10:15,  1.28s/it]
 24%|██▍       | 154/633 [03:29<09:47,  1.23s/it]
 24%|██▍       | 155/633 [03:31<09:35,  1.20s/it]
 25%|██▍       | 156/633 [03:33<11:18,  1.42s/it]
 25%|██▍       | 157/633 [03:34<11:12,  1.41s/it]Saved adapter checkpoint checkpoints/modernbert-large-30k/checkpoint-1of4-step158

 25%|██▍       | 158/633 [03:36<12:49,  1.62s/it]

  0%|          | 0/47 [00:00<?, ?it/s][A

  4%|▍         | 2/47 [00:00<00:07,  5.83it/s][A

  6%|▋         | 3/47 [00:00<00:11,  3.95it/s][A

  9%|▊         | 4/47 [00:01<00:13,  3.27it/s][A

 11%|█         | 5/47 [00:01<00:13,  3.05it/s][A

 13%|█▎        | 6/47 [00:01<00:13,  3.01it/s][A

 15%|█▍        | 7/47 [00:02<00:13,  2.94it/s][A

 17%|█▋        | 8/47 [00:02<00:15,  2.44it/s][A

 19%|█▉        | 9/47 [00:03<00:15,  2.44it/s][A

 21%|██▏       | 10/47 [00:03<00:13,  2.75it/s][A

 23%|██▎       | 11/47 [00:03<00:13,  2.71it/s][A

 26%|██▌       | 12/47 [00:04<00:12,  2.78it/s][A

 28%|██▊       | 13/47 [00:04<00:12,  2.70it/s][A

 30%|██▉       | 14/47 [00:04<00:12,  2.73it/s][A

 32%|███▏      | 15/47 [00:05<00:11,  2.74it/s][A

 34%|███▍      | 16/47 [00:05<00:10,  2.85it/s][A

 36%|███▌      | 17/47 [00:05<00:10,  2.79it/s][A

 38%|███▊      | 18/47 [00:06<00:10,  2.75it/s][A

 40%|████      | 19/47 [00:06<00:09,  2.88it/s][A

 43%|████▎     | 20/47 [00:06<00:09,  2.99it/s][A

 45%|████▍     | 21/47 [00:07<00:10,  2.50it/s][A

 47%|████▋     | 22/47 [00:07<00:09,  2.54it/s][A

 49%|████▉     | 23/47 [00:08<00:08,  2.97it/s][A

 51%|█████     | 24/47 [00:08<00:08,  2.86it/s][A

 53%|█████▎    | 25/47 [00:08<00:07,  2.78it/s][A

 55%|█████▌    | 26/47 [00:09<00:07,  2.72it/s][A

 57%|█████▋    | 27/47 [00:09<00:07,  2.79it/s][A

 60%|█████▉    | 28/47 [00:09<00:06,  2.79it/s][A

 62%|██████▏   | 29/47 [00:10<00:06,  2.77it/s][A

 64%|██████▍   | 30/47 [00:10<00:06,  2.75it/s][A

 66%|██████▌   | 31/47 [00:11<00:05,  2.73it/s][A

 68%|██████▊   | 32/47 [00:11<00:05,  2.69it/s][A

 70%|███████   | 33/47 [00:11<00:05,  2.73it/s][A

 72%|███████▏  | 34/47 [00:12<00:04,  2.95it/s][A

 74%|███████▍  | 35/47 [00:12<00:04,  2.89it/s][A

 77%|███████▋  | 36/47 [00:12<00:03,  2.80it/s][A

 79%|███████▊  | 37/47 [00:13<00:03,  2.74it/s][A

 81%|████████  | 38/47 [00:13<00:03,  2.70it/s][A

 83%|████████▎ | 39/47 [00:14<00:03,  2.33it/s][A

 85%|████████▌ | 40/47 [00:14<00:02,  2.53it/s][A

 87%|████████▋ | 41/47 [00:14<00:02,  2.25it/s][A

 89%|████████▉ | 42/47 [00:15<00:02,  2.27it/s][A

 91%|█████████▏| 43/47 [00:15<00:01,  2.51it/s][A

 94%|█████████▎| 44/47 [00:15<00:01,  2.73it/s][A

 96%|█████████▌| 45/47 [00:16<00:00,  2.60it/s][A

 98%|█████████▊| 46/47 [00:16<00:00,  2.93it/s][A

100%|██████████| 47/47 [00:17<00:00,  2.82it/s][A
                                                 

                                               
[A{'eval_loss': 0.5826137065887451, 'eval_followed_precision': 0.7449768160741885, 'eval_followed_recall': 0.8645739910313901, 'eval_followed_f1': 0.8003320880033209, 'eval_followed_support': 1115, 'eval_violated_precision': 0.8576779026217228, 'eval_violated_recall': 0.5948051948051948, 'eval_violated_f1': 0.7024539877300614, 'eval_violated_support': 770, 'eval_not_applicable_precision': 0.8395904436860068, 'eval_not_applicable_recall': 0.8825112107623319, 'eval_not_applicable_f1': 0.8605159597726278, 'eval_not_applicable_support': 1115, 'eval_accuracy': 0.802, 'eval_macro_f1': 0.7877673451686701, 'eval_runtime': 18.2927, 'eval_samples_per_second': 164.0, 'eval_steps_per_second': 2.569, 'epoch': 0.37}

 25%|██▍       | 158/633 [03:54<12:49,  1.62s/it]

100%|██████████| 47/47 [00:17<00:00,  2.82it/s][A

                                               [A
 25%|██▌       | 159/633 [03:56<55:23,  7.01s/it]
 25%|██▌       | 160/633 [03:57<41:06,  5.21s/it]
                                                 
{'loss': 0.526, 'grad_norm': 2.3687219619750977, 'learning_rate': 0.00015773710482529119, 'epoch': 0.38}

 25%|██▌       | 160/633 [03:57<41:06,  5.21s/it]
 25%|██▌       | 161/633 [03:58<31:41,  4.03s/it]
 26%|██▌       | 162/633 [03:59<24:10,  3.08s/it]
 26%|██▌       | 163/633 [04:00<18:51,  2.41s/it]
 26%|██▌       | 164/633 [04:00<15:16,  1.95s/it]
 26%|██▌       | 165/633 [04:02<13:33,  1.74s/it]
 26%|██▌       | 166/633 [04:03<12:15,  1.58s/it]
 26%|██▋       | 167/633 [04:04<10:35,  1.36s/it]
 27%|██▋       | 168/633 [04:05<09:28,  1.22s/it]
 27%|██▋       | 169/633 [04:06<09:42,  1.26s/it]
 27%|██▋       | 170/633 [04:08<10:52,  1.41s/it]
                                                 
{'loss': 0.4178, 'grad_norm': 2.684997081756592, 'learning_rate': 0.00015440931780366058, 'epoch': 0.4}

 27%|██▋       | 170/633 [04:08<10:52,  1.41s/it]
 27%|██▋       | 171/633 [04:09<10:27,  1.36s/it]
 27%|██▋       | 172/633 [04:10<10:15,  1.33s/it]
 27%|██▋       | 173/633 [04:12<10:20,  1.35s/it]
 27%|██▋       | 174/633 [04:13<10:26,  1.36s/it]
 28%|██▊       | 175/633 [04:14<09:21,  1.23s/it]
 28%|██▊       | 176/633 [04:15<08:41,  1.14s/it]
 28%|██▊       | 177/633 [04:17<10:32,  1.39s/it]
 28%|██▊       | 178/633 [04:18<10:00,  1.32s/it]
 28%|██▊       | 179/633 [04:19<08:58,  1.19s/it]
 28%|██▊       | 180/633 [04:20<08:35,  1.14s/it]
                                                 
{'loss': 0.5192, 'grad_norm': 3.408480167388916, 'learning_rate': 0.00015108153078202997, 'epoch': 0.43}

 28%|██▊       | 180/633 [04:20<08:35,  1.14s/it]
 29%|██▊       | 181/633 [04:21<08:50,  1.17s/it]
 29%|██▉       | 182/633 [04:23<09:19,  1.24s/it]
 29%|██▉       | 183/633 [04:24<09:23,  1.25s/it]
 29%|██▉       | 184/633 [04:25<09:55,  1.33s/it]
 29%|██▉       | 185/633 [04:26<09:01,  1.21s/it]
 29%|██▉       | 186/633 [04:28<09:22,  1.26s/it]
 30%|██▉       | 187/633 [04:29<09:04,  1.22s/it]
 30%|██▉       | 188/633 [04:30<09:24,  1.27s/it]
 30%|██▉       | 189/633 [04:32<09:27,  1.28s/it]
 30%|███       | 190/633 [04:33<10:07,  1.37s/it]
                                                 
{'loss': 0.4658, 'grad_norm': 2.5103650093078613, 'learning_rate': 0.00014775374376039933, 'epoch': 0.45}

 30%|███       | 190/633 [04:33<10:07,  1.37s/it]
 30%|███       | 191/633 [04:35<10:50,  1.47s/it]
 30%|███       | 192/633 [04:36<09:58,  1.36s/it]
 30%|███       | 193/633 [04:37<09:44,  1.33s/it]
 31%|███       | 194/633 [04:38<09:06,  1.24s/it]
 31%|███       | 195/633 [04:39<08:50,  1.21s/it]
 31%|███       | 196/633 [04:41<09:07,  1.25s/it]
 31%|███       | 197/633 [04:42<08:25,  1.16s/it]
 31%|███▏      | 198/633 [04:43<08:53,  1.23s/it]
 31%|███▏      | 199/633 [04:44<09:16,  1.28s/it]
 32%|███▏      | 200/633 [04:46<09:02,  1.25s/it]
                                                 
{'loss': 0.4336, 'grad_norm': 1.899316668510437, 'learning_rate': 0.00014442595673876872, 'epoch': 0.47}

 32%|███▏      | 200/633 [04:46<09:02,  1.25s/it]
 32%|███▏      | 201/633 [04:47<08:59,  1.25s/it]
 32%|███▏      | 202/633 [04:48<08:59,  1.25s/it]
 32%|███▏      | 203/633 [04:49<08:17,  1.16s/it]
 32%|███▏      | 204/633 [04:50<08:28,  1.19s/it]
 32%|███▏      | 205/633 [04:52<10:01,  1.40s/it]
 33%|███▎      | 206/633 [04:54<10:11,  1.43s/it]
 33%|███▎      | 207/633 [04:55<09:01,  1.27s/it]
 33%|███▎      | 208/633 [04:56<08:38,  1.22s/it]
 33%|███▎      | 209/633 [04:57<08:59,  1.27s/it]
 33%|███▎      | 210/633 [04:58<08:18,  1.18s/it]
                                                 
{'loss': 0.4002, 'grad_norm': 2.306647539138794, 'learning_rate': 0.0001410981697171381, 'epoch': 0.5}

 33%|███▎      | 210/633 [04:58<08:18,  1.18s/it]
 33%|███▎      | 211/633 [05:00<09:56,  1.41s/it]
 33%|███▎      | 212/633 [05:02<11:06,  1.58s/it]
 34%|███▎      | 213/633 [05:03<09:27,  1.35s/it]
 34%|███▍      | 214/633 [05:04<09:30,  1.36s/it]
 34%|███▍      | 215/633 [05:06<09:26,  1.35s/it]
 34%|███▍      | 216/633 [05:07<09:13,  1.33s/it]
 34%|███▍      | 217/633 [05:08<09:12,  1.33s/it]
 34%|███▍      | 218/633 [05:09<08:51,  1.28s/it]
 35%|███▍      | 219/633 [05:11<08:56,  1.30s/it]
 35%|███▍      | 220/633 [05:12<08:46,  1.27s/it]
                                                 
{'loss': 0.4077, 'grad_norm': 2.039564609527588, 'learning_rate': 0.0001377703826955075, 'epoch': 0.52}

 35%|███▍      | 220/633 [05:12<08:46,  1.27s/it]
 35%|███▍      | 221/633 [05:13<08:41,  1.27s/it]
 35%|███▌      | 222/633 [05:14<08:24,  1.23s/it]
 35%|███▌      | 223/633 [05:15<07:24,  1.08s/it]
 35%|███▌      | 224/633 [05:17<09:08,  1.34s/it]
 36%|███▌      | 225/633 [05:18<09:00,  1.32s/it]
 36%|███▌      | 226/633 [05:20<09:05,  1.34s/it]
 36%|███▌      | 227/633 [05:21<08:27,  1.25s/it]
 36%|███▌      | 228/633 [05:24<11:49,  1.75s/it]
 36%|███▌      | 229/633 [05:25<11:00,  1.63s/it]
 36%|███▋      | 230/633 [05:26<10:39,  1.59s/it]
                                                 
{'loss': 0.512, 'grad_norm': 3.74591326713562, 'learning_rate': 0.00013444259567387687, 'epoch': 0.55}

 36%|███▋      | 230/633 [05:26<10:39,  1.59s/it]
 36%|███▋      | 231/633 [05:28<11:22,  1.70s/it]
 37%|███▋      | 232/633 [05:30<10:45,  1.61s/it]
 37%|███▋      | 233/633 [05:31<09:31,  1.43s/it]
 37%|███▋      | 234/633 [05:32<08:37,  1.30s/it]
 37%|███▋      | 235/633 [05:33<08:42,  1.31s/it]
 37%|███▋      | 236/633 [05:34<08:34,  1.30s/it]
 37%|███▋      | 237/633 [05:35<07:42,  1.17s/it]
 38%|███▊      | 238/633 [05:38<11:08,  1.69s/it]
 38%|███▊      | 239/633 [05:39<09:41,  1.48s/it]
 38%|███▊      | 240/633 [05:40<09:24,  1.44s/it]
                                                 
{'loss': 0.4139, 'grad_norm': 2.354063034057617, 'learning_rate': 0.00013111480865224626, 'epoch': 0.57}

 38%|███▊      | 240/633 [05:40<09:24,  1.44s/it]
 38%|███▊      | 241/633 [05:42<09:26,  1.45s/it]
 38%|███▊      | 242/633 [05:43<09:30,  1.46s/it]
 38%|███▊      | 243/633 [05:45<09:23,  1.44s/it]
 39%|███▊      | 244/633 [05:46<08:40,  1.34s/it]
 39%|███▊      | 245/633 [05:47<07:34,  1.17s/it]
 39%|███▉      | 246/633 [05:48<07:45,  1.20s/it]
 39%|███▉      | 247/633 [05:49<08:15,  1.28s/it]
 39%|███▉      | 248/633 [05:51<07:54,  1.23s/it]
 39%|███▉      | 249/633 [05:52<07:58,  1.25s/it]
 39%|███▉      | 250/633 [05:53<07:39,  1.20s/it]
                                                 
{'loss': 0.4049, 'grad_norm': 1.6327793598175049, 'learning_rate': 0.00012778702163061565, 'epoch': 0.59}

 39%|███▉      | 250/633 [05:53<07:39,  1.20s/it]
 40%|███▉      | 251/633 [05:54<07:32,  1.18s/it]
 40%|███▉      | 252/633 [05:55<07:13,  1.14s/it]
 40%|███▉      | 253/633 [05:56<07:14,  1.14s/it]
 40%|████      | 254/633 [05:58<08:47,  1.39s/it]
 40%|████      | 255/633 [05:59<08:22,  1.33s/it]
 40%|████      | 256/633 [06:00<07:33,  1.20s/it]
 41%|████      | 257/633 [06:01<07:13,  1.15s/it]
 41%|████      | 258/633 [06:03<07:56,  1.27s/it]
 41%|████      | 259/633 [06:04<08:24,  1.35s/it]
 41%|████      | 260/633 [06:05<07:28,  1.20s/it]
                                                 
{'loss': 0.3744, 'grad_norm': 1.6186234951019287, 'learning_rate': 0.00012445923460898504, 'epoch': 0.62}

 41%|████      | 260/633 [06:05<07:28,  1.20s/it]
 41%|████      | 261/633 [06:07<07:44,  1.25s/it]
 41%|████▏     | 262/633 [06:08<07:14,  1.17s/it]
 42%|████▏     | 263/633 [06:09<07:01,  1.14s/it]
 42%|████▏     | 264/633 [06:10<07:24,  1.20s/it]
 42%|████▏     | 265/633 [06:12<07:51,  1.28s/it]
 42%|████▏     | 266/633 [06:13<07:51,  1.28s/it]
 42%|████▏     | 267/633 [06:14<07:44,  1.27s/it]
 42%|████▏     | 268/633 [06:15<07:42,  1.27s/it]
 42%|████▏     | 269/633 [06:17<07:56,  1.31s/it]
 43%|████▎     | 270/633 [06:18<07:08,  1.18s/it]
                                                 
{'loss': 0.3767, 'grad_norm': 1.6980925798416138, 'learning_rate': 0.00012113144758735442, 'epoch': 0.64}

 43%|████▎     | 270/633 [06:18<07:08,  1.18s/it]
 43%|████▎     | 271/633 [06:19<07:15,  1.20s/it]
 43%|████▎     | 272/633 [06:20<07:15,  1.21s/it]
 43%|████▎     | 273/633 [06:21<06:50,  1.14s/it]
 43%|████▎     | 274/633 [06:24<09:55,  1.66s/it]
 43%|████▎     | 275/633 [06:25<09:08,  1.53s/it]
 44%|████▎     | 276/633 [06:26<08:37,  1.45s/it]
 44%|████▍     | 277/633 [06:28<08:14,  1.39s/it]
 44%|████▍     | 278/633 [06:29<07:28,  1.26s/it]
 44%|████▍     | 279/633 [06:30<07:19,  1.24s/it]
 44%|████▍     | 280/633 [06:32<09:17,  1.58s/it]
                                                 
{'loss': 0.3969, 'grad_norm': 1.7364944219589233, 'learning_rate': 0.00011780366056572381, 'epoch': 0.66}

 44%|████▍     | 280/633 [06:32<09:17,  1.58s/it]
 44%|████▍     | 281/633 [06:33<08:36,  1.47s/it]
 45%|████▍     | 282/633 [06:35<08:24,  1.44s/it]
 45%|████▍     | 283/633 [06:36<08:02,  1.38s/it]
 45%|████▍     | 284/633 [06:37<07:31,  1.29s/it]
 45%|████▌     | 285/633 [06:38<07:15,  1.25s/it]
 45%|████▌     | 286/633 [06:39<06:28,  1.12s/it]
 45%|████▌     | 287/633 [06:40<06:10,  1.07s/it]
 45%|████▌     | 288/633 [06:41<06:51,  1.19s/it]
 46%|████▌     | 289/633 [06:43<06:54,  1.21s/it]
 46%|████▌     | 290/633 [06:44<06:48,  1.19s/it]
                                                 
{'loss': 0.3923, 'grad_norm': 2.798112630844116, 'learning_rate': 0.0001144758735440932, 'epoch': 0.69}

 46%|████▌     | 290/633 [06:44<06:48,  1.19s/it]
 46%|████▌     | 291/633 [06:45<06:51,  1.20s/it]
 46%|████▌     | 292/633 [06:46<06:05,  1.07s/it]
 46%|████▋     | 293/633 [06:47<05:53,  1.04s/it]
 46%|████▋     | 294/633 [06:49<07:21,  1.30s/it]
 47%|████▋     | 295/633 [06:50<07:59,  1.42s/it]
 47%|████▋     | 296/633 [06:52<08:34,  1.53s/it]
 47%|████▋     | 297/633 [06:54<08:47,  1.57s/it]
 47%|████▋     | 298/633 [06:55<08:03,  1.44s/it]
 47%|████▋     | 299/633 [06:56<07:41,  1.38s/it]
 47%|████▋     | 300/633 [06:57<07:02,  1.27s/it]
                                                 
{'loss': 0.3444, 'grad_norm': 1.6849703788757324, 'learning_rate': 0.00011114808652246256, 'epoch': 0.71}

 47%|████▋     | 300/633 [06:57<07:02,  1.27s/it]
 48%|████▊     | 301/633 [06:59<07:14,  1.31s/it]
 48%|████▊     | 302/633 [07:00<07:16,  1.32s/it]
 48%|████▊     | 303/633 [07:01<07:29,  1.36s/it]
 48%|████▊     | 304/633 [07:03<07:13,  1.32s/it]
 48%|████▊     | 305/633 [07:04<06:54,  1.26s/it]
 48%|████▊     | 306/633 [07:06<07:40,  1.41s/it]
 48%|████▊     | 307/633 [07:07<07:23,  1.36s/it]
 49%|████▊     | 308/633 [07:08<07:20,  1.36s/it]
 49%|████▉     | 309/633 [07:10<07:20,  1.36s/it]
 49%|████▉     | 310/633 [07:11<07:10,  1.33s/it]
                                                 
{'loss': 0.4533, 'grad_norm': 1.760393500328064, 'learning_rate': 0.00010782029950083194, 'epoch': 0.73}

 49%|████▉     | 310/633 [07:11<07:10,  1.33s/it]
 49%|████▉     | 311/633 [07:13<08:05,  1.51s/it]
 49%|████▉     | 312/633 [07:15<08:45,  1.64s/it]
 49%|████▉     | 313/633 [07:16<08:02,  1.51s/it]
 50%|████▉     | 314/633 [07:17<07:06,  1.34s/it]
 50%|████▉     | 315/633 [07:18<07:10,  1.35s/it]Saved adapter checkpoint checkpoints/modernbert-large-30k/checkpoint-2of4-step316

 50%|████▉     | 316/633 [07:20<08:26,  1.60s/it]

  0%|          | 0/47 [00:00<?, ?it/s][A

  4%|▍         | 2/47 [00:00<00:07,  5.84it/s][A

  6%|▋         | 3/47 [00:00<00:11,  3.93it/s][A

  9%|▊         | 4/47 [00:01<00:13,  3.28it/s][A

 11%|█         | 5/47 [00:01<00:13,  3.06it/s][A

 13%|█▎        | 6/47 [00:01<00:13,  3.08it/s][A

 15%|█▍        | 7/47 [00:02<00:13,  2.93it/s][A

 17%|█▋        | 8/47 [00:02<00:16,  2.38it/s][A

 19%|█▉        | 9/47 [00:03<00:15,  2.42it/s][A

 21%|██▏       | 10/47 [00:03<00:13,  2.69it/s][A

 23%|██▎       | 11/47 [00:03<00:13,  2.69it/s][A

 26%|██▌       | 12/47 [00:04<00:12,  2.76it/s][A

 28%|██▊       | 13/47 [00:04<00:12,  2.69it/s][A

 30%|██▉       | 14/47 [00:04<00:12,  2.74it/s][A

 32%|███▏      | 15/47 [00:05<00:11,  2.75it/s][A

 34%|███▍      | 16/47 [00:05<00:10,  2.85it/s][A

 36%|███▌      | 17/47 [00:05<00:10,  2.86it/s][A

 38%|███▊      | 18/47 [00:06<00:10,  2.76it/s][A

 40%|████      | 19/47 [00:06<00:09,  2.92it/s][A

 43%|████▎     | 20/47 [00:06<00:08,  3.04it/s][A

 45%|████▍     | 21/47 [00:07<00:10,  2.53it/s][A

 47%|████▋     | 22/47 [00:07<00:09,  2.57it/s][A

 49%|████▉     | 23/47 [00:08<00:08,  2.99it/s][A

 51%|█████     | 24/47 [00:08<00:08,  2.86it/s][A

 53%|█████▎    | 25/47 [00:08<00:07,  2.80it/s][A

 55%|█████▌    | 26/47 [00:09<00:07,  2.73it/s][A

 57%|█████▋    | 27/47 [00:09<00:07,  2.79it/s][A

 60%|█████▉    | 28/47 [00:09<00:06,  2.80it/s][A

 62%|██████▏   | 29/47 [00:10<00:06,  2.77it/s][A

 64%|██████▍   | 30/47 [00:10<00:06,  2.74it/s][A

 66%|██████▌   | 31/47 [00:10<00:05,  2.72it/s][A

 68%|██████▊   | 32/47 [00:11<00:05,  2.70it/s][A

 70%|███████   | 33/47 [00:11<00:05,  2.73it/s][A

 72%|███████▏  | 34/47 [00:12<00:04,  2.93it/s][A

 74%|███████▍  | 35/47 [00:12<00:04,  2.87it/s][A

 77%|███████▋  | 36/47 [00:12<00:03,  2.78it/s][A

 79%|███████▊  | 37/47 [00:13<00:03,  2.72it/s][A

 81%|████████  | 38/47 [00:13<00:03,  2.69it/s][A

 83%|████████▎ | 39/47 [00:14<00:03,  2.32it/s][A

 85%|████████▌ | 40/47 [00:14<00:02,  2.52it/s][A

 87%|████████▋ | 41/47 [00:14<00:02,  2.25it/s][A

 89%|████████▉ | 42/47 [00:15<00:02,  2.27it/s][A

 91%|█████████▏| 43/47 [00:15<00:01,  2.53it/s][A

 94%|█████████▎| 44/47 [00:15<00:01,  2.74it/s][A

 96%|█████████▌| 45/47 [00:16<00:00,  2.60it/s][A

 98%|█████████▊| 46/47 [00:16<00:00,  2.93it/s][A

100%|██████████| 47/47 [00:17<00:00,  2.86it/s][A
                                                 

                                               
[A{'eval_loss': 0.4957300126552582, 'eval_followed_precision': 0.7516828721017202, 'eval_followed_recall': 0.9013452914798207, 'eval_followed_f1': 0.8197389885807504, 'eval_followed_support': 1115, 'eval_violated_precision': 0.7891061452513967, 'eval_violated_recall': 0.7337662337662337, 'eval_violated_f1': 0.7604306864064603, 'eval_violated_support': 770, 'eval_not_applicable_precision': 0.9186906019007391, 'eval_not_applicable_recall': 0.7802690582959642, 'eval_not_applicable_f1': 0.8438409311348206, 'eval_not_applicable_support': 1115, 'eval_accuracy': 0.8133333333333334, 'eval_macro_f1': 0.8080035353740104, 'eval_runtime': 17.8463, 'eval_samples_per_second': 168.102, 'eval_steps_per_second': 2.634, 'epoch': 0.75}

 50%|████▉     | 316/633 [07:38<08:26,  1.60s/it]

100%|██████████| 47/47 [00:17<00:00,  2.86it/s][A

                                               [A
 50%|█████     | 317/633 [07:39<36:05,  6.85s/it]
 50%|█████     | 318/633 [07:41<27:50,  5.30s/it]
 50%|█████     | 319/633 [07:43<21:38,  4.13s/it]
 51%|█████     | 320/633 [07:44<17:01,  3.26s/it]
                                                 
{'loss': 0.3899, 'grad_norm': 1.97598135471344, 'learning_rate': 0.00010449251247920133, 'epoch': 0.76}

 51%|█████     | 320/633 [07:44<17:01,  3.26s/it]
 51%|█████     | 321/633 [07:45<13:34,  2.61s/it]
 51%|█████     | 322/633 [07:46<11:32,  2.23s/it]
 51%|█████     | 323/633 [07:48<10:55,  2.11s/it]
 51%|█████     | 324/633 [07:49<09:31,  1.85s/it]
 51%|█████▏    | 325/633 [07:51<09:13,  1.80s/it]
 52%|█████▏    | 326/633 [07:52<08:08,  1.59s/it]
 52%|█████▏    | 327/633 [07:53<07:01,  1.38s/it]
 52%|█████▏    | 328/633 [07:55<07:29,  1.48s/it]
 52%|█████▏    | 329/633 [07:56<07:06,  1.40s/it]
 52%|█████▏    | 330/633 [07:57<06:52,  1.36s/it]
                                                 
{'loss': 0.3805, 'grad_norm': 1.554028034210205, 'learning_rate': 0.00010116472545757072, 'epoch': 0.78}

 52%|█████▏    | 330/633 [07:57<06:52,  1.36s/it]
 52%|█████▏    | 331/633 [08:00<09:01,  1.79s/it]
 52%|█████▏    | 332/633 [08:01<08:00,  1.60s/it]
 53%|█████▎    | 333/633 [08:02<07:12,  1.44s/it]
 53%|█████▎    | 334/633 [08:03<06:22,  1.28s/it]
 53%|█████▎    | 335/633 [08:04<05:45,  1.16s/it]
 53%|█████▎    | 336/633 [08:05<05:40,  1.15s/it]
 53%|█████▎    | 337/633 [08:07<06:46,  1.37s/it]
 53%|█████▎    | 338/633 [08:09<07:33,  1.54s/it]
 54%|█████▎    | 339/633 [08:10<07:10,  1.46s/it]
 54%|█████▎    | 340/633 [08:13<09:12,  1.88s/it]
                                                 
{'loss': 0.3417, 'grad_norm': 2.20367169380188, 'learning_rate': 9.783693843594011e-05, 'epoch': 0.81}

 54%|█████▎    | 340/633 [08:13<09:12,  1.88s/it]
 54%|█████▍    | 341/633 [08:14<08:10,  1.68s/it]
 54%|█████▍    | 342/633 [08:15<07:19,  1.51s/it]
 54%|█████▍    | 343/633 [08:17<07:07,  1.47s/it]
 54%|█████▍    | 344/633 [08:18<06:44,  1.40s/it]
 55%|█████▍    | 345/633 [08:20<08:09,  1.70s/it]
 55%|█████▍    | 346/633 [08:22<08:25,  1.76s/it]
 55%|█████▍    | 347/633 [08:24<07:37,  1.60s/it]
 55%|█████▍    | 348/633 [08:25<07:25,  1.56s/it]
 55%|█████▌    | 349/633 [08:26<06:58,  1.47s/it]
 55%|█████▌    | 350/633 [08:28<06:36,  1.40s/it]
                                                 
{'loss': 0.3207, 'grad_norm': 2.0488440990448, 'learning_rate': 9.450915141430949e-05, 'epoch': 0.83}

 55%|█████▌    | 350/633 [08:28<06:36,  1.40s/it]
 55%|█████▌    | 351/633 [08:30<07:59,  1.70s/it]
 56%|█████▌    | 352/633 [08:31<07:08,  1.52s/it]
 56%|█████▌    | 353/633 [08:32<06:53,  1.48s/it]
 56%|█████▌    | 354/633 [08:33<06:20,  1.36s/it]
 56%|█████▌    | 355/633 [08:35<05:55,  1.28s/it]
 56%|█████▌    | 356/633 [08:36<06:04,  1.32s/it]
 56%|█████▋    | 357/633 [08:37<05:25,  1.18s/it]
 57%|█████▋    | 358/633 [08:38<05:25,  1.18s/it]
 57%|█████▋    | 359/633 [08:39<04:56,  1.08s/it]
 57%|█████▋    | 360/633 [08:40<05:07,  1.13s/it]
                                                 
{'loss': 0.3444, 'grad_norm': 1.5197449922561646, 'learning_rate': 9.118136439267887e-05, 'epoch': 0.85}

 57%|█████▋    | 360/633 [08:40<05:07,  1.13s/it]
 57%|█████▋    | 361/633 [08:41<05:18,  1.17s/it]
 57%|█████▋    | 362/633 [08:43<05:15,  1.16s/it]
 57%|█████▋    | 363/633 [08:44<05:10,  1.15s/it]
 58%|█████▊    | 364/633 [08:45<04:47,  1.07s/it]
 58%|█████▊    | 365/633 [08:46<05:20,  1.20s/it]
 58%|█████▊    | 366/633 [08:47<05:35,  1.26s/it]
 58%|█████▊    | 367/633 [08:49<05:57,  1.35s/it]
 58%|█████▊    | 368/633 [08:50<05:38,  1.28s/it]
 58%|█████▊    | 369/633 [08:51<05:34,  1.27s/it]
 58%|█████▊    | 370/633 [08:52<05:11,  1.18s/it]
                                                 
{'loss': 0.3945, 'grad_norm': 1.9636036157608032, 'learning_rate': 8.785357737104826e-05, 'epoch': 0.88}

 58%|█████▊    | 370/633 [08:52<05:11,  1.18s/it]
 59%|█████▊    | 371/633 [08:54<05:12,  1.19s/it]
 59%|█████▉    | 372/633 [08:55<05:17,  1.21s/it]
 59%|█████▉    | 373/633 [08:57<05:53,  1.36s/it]
 59%|█████▉    | 374/633 [08:57<05:19,  1.23s/it]
 59%|█████▉    | 375/633 [09:00<07:33,  1.76s/it]
 59%|█████▉    | 376/633 [09:02<07:33,  1.77s/it]
 60%|█████▉    | 377/633 [09:04<07:44,  1.81s/it]
 60%|█████▉    | 378/633 [09:06<07:08,  1.68s/it]
 60%|█████▉    | 379/633 [09:06<06:01,  1.42s/it]
 60%|██████    | 380/633 [09:07<05:36,  1.33s/it]
                                                 
{'loss': 0.3453, 'grad_norm': 1.2806130647659302, 'learning_rate': 8.452579034941765e-05, 'epoch': 0.9}

 60%|██████    | 380/633 [09:07<05:36,  1.33s/it]
 60%|██████    | 381/633 [09:09<05:38,  1.34s/it]
 60%|██████    | 382/633 [09:11<06:06,  1.46s/it]
 61%|██████    | 383/633 [09:12<05:44,  1.38s/it]
 61%|██████    | 384/633 [09:13<05:44,  1.39s/it]
 61%|██████    | 385/633 [09:14<05:39,  1.37s/it]
 61%|██████    | 386/633 [09:15<05:00,  1.22s/it]
 61%|██████    | 387/633 [09:17<05:59,  1.46s/it]
 61%|██████▏   | 388/633 [09:19<05:58,  1.46s/it]
 61%|██████▏   | 389/633 [09:20<05:42,  1.40s/it]
 62%|██████▏   | 390/633 [09:21<05:28,  1.35s/it]
                                                 
{'loss': 0.301, 'grad_norm': 1.7552975416183472, 'learning_rate': 8.119800332778702e-05, 'epoch': 0.92}

 62%|██████▏   | 390/633 [09:21<05:28,  1.35s/it]
 62%|██████▏   | 391/633 [09:23<05:20,  1.33s/it]
 62%|██████▏   | 392/633 [09:24<05:10,  1.29s/it]
 62%|██████▏   | 393/633 [09:25<04:45,  1.19s/it]
 62%|██████▏   | 394/633 [09:26<04:14,  1.06s/it]
 62%|██████▏   | 395/633 [09:28<06:29,  1.64s/it]
 63%|██████▎   | 396/633 [09:31<07:22,  1.87s/it]
 63%|██████▎   | 397/633 [09:32<06:12,  1.58s/it]
 63%|██████▎   | 398/633 [09:34<06:47,  1.73s/it]
 63%|██████▎   | 399/633 [09:35<06:28,  1.66s/it]
 63%|██████▎   | 400/633 [09:37<06:29,  1.67s/it]
                                                 
{'loss': 0.2897, 'grad_norm': 1.1469783782958984, 'learning_rate': 7.78702163061564e-05, 'epoch': 0.95}

 63%|██████▎   | 400/633 [09:37<06:29,  1.67s/it]
 63%|██████▎   | 401/633 [09:38<05:49,  1.51s/it]
 64%|██████▎   | 402/633 [09:40<05:36,  1.46s/it]
 64%|██████▎   | 403/633 [09:42<06:15,  1.63s/it]
 64%|██████▍   | 404/633 [09:43<05:51,  1.53s/it]
 64%|██████▍   | 405/633 [09:44<05:22,  1.41s/it]
 64%|██████▍   | 406/633 [09:46<05:38,  1.49s/it]
 64%|██████▍   | 407/633 [09:47<04:52,  1.30s/it]
 64%|██████▍   | 408/633 [09:48<04:34,  1.22s/it]
 65%|██████▍   | 409/633 [09:48<03:59,  1.07s/it]
 65%|██████▍   | 410/633 [09:50<04:16,  1.15s/it]
                                                 
{'loss': 0.3556, 'grad_norm': 1.671075701713562, 'learning_rate': 7.454242928452579e-05, 'epoch': 0.97}

 65%|██████▍   | 410/633 [09:50<04:16,  1.15s/it]
 65%|██████▍   | 411/633 [09:51<04:19,  1.17s/it]
 65%|██████▌   | 412/633 [09:53<05:06,  1.39s/it]
 65%|██████▌   | 413/633 [09:54<04:35,  1.25s/it]
 65%|██████▌   | 414/633 [09:56<05:17,  1.45s/it]
 66%|██████▌   | 415/633 [09:58<05:47,  1.59s/it]
 66%|██████▌   | 416/633 [09:58<05:04,  1.40s/it]
 66%|██████▌   | 417/633 [10:00<04:53,  1.36s/it]
 66%|██████▌   | 418/633 [10:01<04:59,  1.39s/it]
 66%|██████▌   | 419/633 [10:03<05:22,  1.51s/it]
 66%|██████▋   | 420/633 [10:04<04:59,  1.40s/it]
                                                 
{'loss': 0.358, 'grad_norm': 2.946380615234375, 'learning_rate': 7.121464226289518e-05, 'epoch': 1.0}

 66%|██████▋   | 420/633 [10:04<04:59,  1.40s/it]
 67%|██████▋   | 421/633 [10:06<05:20,  1.51s/it]
 67%|██████▋   | 422/633 [10:07<04:52,  1.39s/it]
 67%|██████▋   | 423/633 [10:09<05:03,  1.45s/it]
 67%|██████▋   | 424/633 [10:10<04:42,  1.35s/it]
 67%|██████▋   | 425/633 [10:11<04:07,  1.19s/it]
 67%|██████▋   | 426/633 [10:11<03:38,  1.06s/it]
 67%|██████▋   | 427/633 [10:13<03:59,  1.16s/it]
 68%|██████▊   | 428/633 [10:14<03:54,  1.14s/it]
 68%|██████▊   | 429/633 [10:15<03:57,  1.16s/it]
 68%|██████▊   | 430/633 [10:16<03:42,  1.10s/it]
                                                 
{'loss': 0.2373, 'grad_norm': 1.1884970664978027, 'learning_rate': 6.788685524126456e-05, 'epoch': 1.02}

 68%|██████▊   | 430/633 [10:16<03:42,  1.10s/it]
 68%|██████▊   | 431/633 [10:17<04:08,  1.23s/it]
 68%|██████▊   | 432/633 [10:19<03:59,  1.19s/it]
 68%|██████▊   | 433/633 [10:20<04:01,  1.21s/it]
 69%|██████▊   | 434/633 [10:22<04:29,  1.36s/it]
 69%|██████▊   | 435/633 [10:23<04:22,  1.32s/it]
 69%|██████▉   | 436/633 [10:24<04:16,  1.30s/it]
 69%|██████▉   | 437/633 [10:25<04:23,  1.35s/it]
 69%|██████▉   | 438/633 [10:27<04:32,  1.40s/it]
 69%|██████▉   | 439/633 [10:29<05:30,  1.71s/it]
 70%|██████▉   | 440/633 [10:31<05:09,  1.60s/it]
                                                 
{'loss': 0.2916, 'grad_norm': 2.3900575637817383, 'learning_rate': 6.455906821963394e-05, 'epoch': 1.04}

 70%|██████▉   | 440/633 [10:31<05:09,  1.60s/it]
 70%|██████▉   | 441/633 [10:32<04:19,  1.35s/it]
 70%|██████▉   | 442/633 [10:32<03:49,  1.20s/it]
 70%|██████▉   | 443/633 [10:34<04:36,  1.45s/it]
 70%|███████   | 444/633 [10:36<04:23,  1.39s/it]
 70%|███████   | 445/633 [10:37<04:06,  1.31s/it]
 70%|███████   | 446/633 [10:38<03:54,  1.26s/it]
 71%|███████   | 447/633 [10:39<03:44,  1.21s/it]
 71%|███████   | 448/633 [10:41<04:24,  1.43s/it]
 71%|███████   | 449/633 [10:42<04:16,  1.39s/it]
 71%|███████   | 450/633 [10:44<04:11,  1.37s/it]
                                                 
{'loss': 0.1917, 'grad_norm': 1.591640830039978, 'learning_rate': 6.123128119800333e-05, 'epoch': 1.07}

 71%|███████   | 450/633 [10:44<04:11,  1.37s/it]
 71%|███████   | 451/633 [10:45<03:42,  1.22s/it]
 71%|███████▏  | 452/633 [10:46<04:22,  1.45s/it]
 72%|███████▏  | 453/633 [10:48<04:03,  1.35s/it]
 72%|███████▏  | 454/633 [10:49<04:07,  1.38s/it]
 72%|███████▏  | 455/633 [10:50<03:42,  1.25s/it]
 72%|███████▏  | 456/633 [10:51<03:43,  1.26s/it]
 72%|███████▏  | 457/633 [10:53<03:41,  1.26s/it]
 72%|███████▏  | 458/633 [10:54<03:25,  1.18s/it]
 73%|███████▎  | 459/633 [10:56<04:58,  1.71s/it]
 73%|███████▎  | 460/633 [10:58<04:24,  1.53s/it]
                                                 
{'loss': 0.2797, 'grad_norm': 1.784266710281372, 'learning_rate': 5.790349417637272e-05, 'epoch': 1.09}

 73%|███████▎  | 460/633 [10:58<04:24,  1.53s/it]
 73%|███████▎  | 461/633 [10:59<04:16,  1.49s/it]
 73%|███████▎  | 462/633 [11:00<04:01,  1.41s/it]
 73%|███████▎  | 463/633 [11:01<03:29,  1.23s/it]
 73%|███████▎  | 464/633 [11:02<03:30,  1.24s/it]
 73%|███████▎  | 465/633 [11:03<03:08,  1.12s/it]
 74%|███████▎  | 466/633 [11:04<03:08,  1.13s/it]
 74%|███████▍  | 467/633 [11:05<03:06,  1.12s/it]
 74%|███████▍  | 468/633 [11:07<03:44,  1.36s/it]
 74%|███████▍  | 469/633 [11:09<04:03,  1.49s/it]
 74%|███████▍  | 470/633 [11:10<03:49,  1.41s/it]
                                                 
{'loss': 0.2395, 'grad_norm': 1.4155722856521606, 'learning_rate': 5.45757071547421e-05, 'epoch': 1.11}

 74%|███████▍  | 470/633 [11:10<03:49,  1.41s/it]
 74%|███████▍  | 471/633 [11:11<03:16,  1.21s/it]
 75%|███████▍  | 472/633 [11:12<03:20,  1.24s/it]
 75%|███████▍  | 473/633 [11:14<03:19,  1.25s/it]
 75%|███████▍  | 474/633 [11:15<03:23,  1.28s/it]Saved adapter checkpoint checkpoints/modernbert-large-30k/checkpoint-3of4-step475

 75%|███████▌  | 475/633 [11:18<04:25,  1.68s/it]

  0%|          | 0/47 [00:00<?, ?it/s][A

  4%|▍         | 2/47 [00:00<00:08,  5.49it/s][A

  6%|▋         | 3/47 [00:00<00:11,  3.86it/s][A

  9%|▊         | 4/47 [00:01<00:13,  3.24it/s][A

 11%|█         | 5/47 [00:01<00:13,  3.03it/s][A

 13%|█▎        | 6/47 [00:01<00:13,  3.06it/s][A

 15%|█▍        | 7/47 [00:02<00:13,  2.95it/s][A

 17%|█▋        | 8/47 [00:02<00:16,  2.42it/s][A

 19%|█▉        | 9/47 [00:03<00:15,  2.41it/s][A

 21%|██▏       | 10/47 [00:03<00:13,  2.73it/s][A

 23%|██▎       | 11/47 [00:03<00:13,  2.72it/s][A

 26%|██▌       | 12/47 [00:04<00:12,  2.76it/s][A

 28%|██▊       | 13/47 [00:04<00:12,  2.69it/s][A

 30%|██▉       | 14/47 [00:04<00:12,  2.73it/s][A

 32%|███▏      | 15/47 [00:05<00:11,  2.76it/s][A

 34%|███▍      | 16/47 [00:05<00:10,  2.86it/s][A

 36%|███▌      | 17/47 [00:05<00:10,  2.84it/s][A

 38%|███▊      | 18/47 [00:06<00:10,  2.77it/s][A

 40%|████      | 19/47 [00:06<00:09,  2.93it/s][A

 43%|████▎     | 20/47 [00:06<00:08,  3.04it/s][A

 45%|████▍     | 21/47 [00:07<00:10,  2.50it/s][A

 47%|████▋     | 22/47 [00:07<00:09,  2.54it/s][A

 49%|████▉     | 23/47 [00:08<00:08,  2.97it/s][A

 51%|█████     | 24/47 [00:08<00:08,  2.87it/s][A

 53%|█████▎    | 25/47 [00:08<00:07,  2.79it/s][A

 55%|█████▌    | 26/47 [00:09<00:07,  2.72it/s][A

 57%|█████▋    | 27/47 [00:09<00:07,  2.78it/s][A

 60%|█████▉    | 28/47 [00:09<00:06,  2.79it/s][A

 62%|██████▏   | 29/47 [00:10<00:06,  2.76it/s][A

 64%|██████▍   | 30/47 [00:10<00:06,  2.75it/s][A

 66%|██████▌   | 31/47 [00:11<00:05,  2.72it/s][A

 68%|██████▊   | 32/47 [00:11<00:05,  2.69it/s][A

 70%|███████   | 33/47 [00:11<00:05,  2.73it/s][A

 72%|███████▏  | 34/47 [00:12<00:04,  2.95it/s][A

 74%|███████▍  | 35/47 [00:12<00:04,  2.88it/s][A

 77%|███████▋  | 36/47 [00:12<00:03,  2.80it/s][A

 79%|███████▊  | 37/47 [00:13<00:03,  2.73it/s][A

 81%|████████  | 38/47 [00:13<00:03,  2.69it/s][A

 83%|████████▎ | 39/47 [00:14<00:03,  2.31it/s][A

 85%|████████▌ | 40/47 [00:14<00:02,  2.51it/s][A

 87%|████████▋ | 41/47 [00:14<00:02,  2.24it/s][A

 89%|████████▉ | 42/47 [00:15<00:02,  2.26it/s][A

 91%|█████████▏| 43/47 [00:15<00:01,  2.50it/s][A

 94%|█████████▎| 44/47 [00:16<00:01,  2.71it/s][A

 96%|█████████▌| 45/47 [00:16<00:00,  2.59it/s][A

 98%|█████████▊| 46/47 [00:16<00:00,  2.92it/s][A

100%|██████████| 47/47 [00:17<00:00,  2.79it/s][A
                                                 

                                               
[A{'eval_loss': 0.5351470112800598, 'eval_followed_precision': 0.8137432188065099, 'eval_followed_recall': 0.8071748878923767, 'eval_followed_f1': 0.8104457451598379, 'eval_followed_support': 1115, 'eval_violated_precision': 0.7378640776699029, 'eval_violated_recall': 0.7896103896103897, 'eval_violated_f1': 0.7628607277289837, 'eval_violated_support': 770, 'eval_not_applicable_precision': 0.8803738317757009, 'eval_not_applicable_recall': 0.8448430493273542, 'eval_not_applicable_f1': 0.8622425629290618, 'eval_not_applicable_support': 1115, 'eval_accuracy': 0.8166666666666667, 'eval_macro_f1': 0.8118496786059611, 'eval_runtime': 17.889, 'eval_samples_per_second': 167.701, 'eval_steps_per_second': 2.627, 'epoch': 1.13}

 75%|███████▌  | 475/633 [11:35<04:25,  1.68s/it]

100%|██████████| 47/47 [00:17<00:00,  2.79it/s][A

                                               [A
 75%|███████▌  | 476/633 [11:37<18:12,  6.96s/it]
 75%|███████▌  | 477/633 [11:39<14:31,  5.59s/it]
 76%|███████▌  | 478/633 [11:41<11:08,  4.31s/it]
 76%|███████▌  | 479/633 [11:42<08:40,  3.38s/it]
 76%|███████▌  | 480/633 [11:43<07:00,  2.75s/it]
                                                 
{'loss': 0.2719, 'grad_norm': 1.556265950202942, 'learning_rate': 5.124792013311148e-05, 'epoch': 1.14}

 76%|███████▌  | 480/633 [11:43<07:00,  2.75s/it]
 76%|███████▌  | 481/633 [11:44<05:49,  2.30s/it]
 76%|███████▌  | 482/633 [11:46<04:59,  1.98s/it]
 76%|███████▋  | 483/633 [11:47<04:24,  1.76s/it]
 76%|███████▋  | 484/633 [11:48<04:05,  1.65s/it]
 77%|███████▋  | 485/633 [11:50<04:05,  1.66s/it]
 77%|███████▋  | 486/633 [11:51<03:40,  1.50s/it]
 77%|███████▋  | 487/633 [11:53<03:46,  1.55s/it]
 77%|███████▋  | 488/633 [11:54<03:20,  1.38s/it]
 77%|███████▋  | 489/633 [11:55<03:16,  1.36s/it]
 77%|███████▋  | 490/633 [11:57<03:33,  1.49s/it]
                                                 
{'loss': 0.2484, 'grad_norm': 1.7409478425979614, 'learning_rate': 4.792013311148087e-05, 'epoch': 1.16}

 77%|███████▋  | 490/633 [11:57<03:33,  1.49s/it]
 78%|███████▊  | 491/633 [11:58<03:28,  1.47s/it]
 78%|███████▊  | 492/633 [11:59<03:19,  1.41s/it]
 78%|███████▊  | 493/633 [12:01<03:29,  1.50s/it]
 78%|███████▊  | 494/633 [12:02<03:12,  1.39s/it]
 78%|███████▊  | 495/633 [12:03<02:58,  1.29s/it]
 78%|███████▊  | 496/633 [12:05<03:06,  1.36s/it]
 79%|███████▊  | 497/633 [12:07<03:30,  1.55s/it]
 79%|███████▊  | 498/633 [12:08<03:19,  1.48s/it]
 79%|███████▉  | 499/633 [12:10<03:12,  1.43s/it]
 79%|███████▉  | 500/633 [12:11<02:55,  1.32s/it]
                                                 
{'loss': 0.2578, 'grad_norm': 1.437913417816162, 'learning_rate': 4.459234608985025e-05, 'epoch': 1.18}

 79%|███████▉  | 500/633 [12:11<02:55,  1.32s/it]
 79%|███████▉  | 501/633 [12:12<02:58,  1.35s/it]
 79%|███████▉  | 502/633 [12:14<03:18,  1.51s/it]
 79%|███████▉  | 503/633 [12:16<03:23,  1.56s/it]
 80%|███████▉  | 504/633 [12:17<03:05,  1.44s/it]
 80%|███████▉  | 505/633 [12:18<03:01,  1.42s/it]
 80%|███████▉  | 506/633 [12:20<03:01,  1.43s/it]
 80%|████████  | 507/633 [12:21<03:09,  1.50s/it]
 80%|████████  | 508/633 [12:22<02:44,  1.32s/it]
 80%|████████  | 509/633 [12:24<02:46,  1.34s/it]
 81%|████████  | 510/633 [12:25<02:41,  1.31s/it]
                                                 
{'loss': 0.2484, 'grad_norm': 1.512278437614441, 'learning_rate': 4.126455906821964e-05, 'epoch': 1.21}

 81%|████████  | 510/633 [12:25<02:41,  1.31s/it]
 81%|████████  | 511/633 [12:26<02:31,  1.24s/it]
 81%|████████  | 512/633 [12:28<02:56,  1.46s/it]
 81%|████████  | 513/633 [12:29<02:34,  1.29s/it]
 81%|████████  | 514/633 [12:30<02:28,  1.25s/it]
 81%|████████▏ | 515/633 [12:31<02:17,  1.16s/it]
 82%|████████▏ | 516/633 [12:32<02:11,  1.12s/it]
 82%|████████▏ | 517/633 [12:33<02:05,  1.08s/it]
 82%|████████▏ | 518/633 [12:34<02:14,  1.17s/it]
 82%|████████▏ | 519/633 [12:36<02:19,  1.22s/it]
 82%|████████▏ | 520/633 [12:37<02:12,  1.17s/it]
                                                 
{'loss': 0.2192, 'grad_norm': 1.4562331438064575, 'learning_rate': 3.793677204658902e-05, 'epoch': 1.23}

 82%|████████▏ | 520/633 [12:37<02:12,  1.17s/it]
 82%|████████▏ | 521/633 [12:39<03:08,  1.68s/it]
 82%|████████▏ | 522/633 [12:41<02:51,  1.55s/it]
 83%|████████▎ | 523/633 [12:42<02:39,  1.45s/it]
 83%|████████▎ | 524/633 [12:43<02:32,  1.40s/it]
 83%|████████▎ | 525/633 [12:45<02:48,  1.56s/it]
 83%|████████▎ | 526/633 [12:46<02:37,  1.47s/it]
 83%|████████▎ | 527/633 [12:49<03:22,  1.91s/it]
 83%|████████▎ | 528/633 [12:50<02:57,  1.69s/it]
 84%|████████▎ | 529/633 [12:52<02:55,  1.69s/it]
 84%|████████▎ | 530/633 [12:53<02:32,  1.48s/it]
                                                 
{'loss': 0.2316, 'grad_norm': 1.9383567571640015, 'learning_rate': 3.460898502495841e-05, 'epoch': 1.26}

 84%|████████▎ | 530/633 [12:53<02:32,  1.48s/it]
 84%|████████▍ | 531/633 [12:55<02:49,  1.67s/it]
 84%|████████▍ | 532/633 [12:56<02:31,  1.50s/it]
 84%|████████▍ | 533/633 [12:58<02:25,  1.46s/it]
 84%|████████▍ | 534/633 [12:59<02:11,  1.33s/it]
 85%|████████▍ | 535/633 [13:01<02:25,  1.48s/it]
 85%|████████▍ | 536/633 [13:02<02:20,  1.45s/it]
 85%|████████▍ | 537/633 [13:03<02:09,  1.35s/it]
 85%|████████▍ | 538/633 [13:04<02:07,  1.35s/it]
 85%|████████▌ | 539/633 [13:05<01:55,  1.22s/it]
 85%|████████▌ | 540/633 [13:07<02:10,  1.40s/it]
                                                 
{'loss': 0.2508, 'grad_norm': 2.012298822402954, 'learning_rate': 3.128119800332779e-05, 'epoch': 1.28}

 85%|████████▌ | 540/633 [13:07<02:10,  1.40s/it]
 85%|████████▌ | 541/633 [13:09<02:18,  1.51s/it]
 86%|████████▌ | 542/633 [13:10<02:10,  1.44s/it]
 86%|████████▌ | 543/633 [13:12<02:17,  1.53s/it]
 86%|████████▌ | 544/633 [13:13<02:07,  1.44s/it]
 86%|████████▌ | 545/633 [13:15<02:07,  1.45s/it]
 86%|████████▋ | 546/633 [13:16<02:05,  1.44s/it]
 86%|████████▋ | 547/633 [13:18<02:04,  1.45s/it]
 87%|████████▋ | 548/633 [13:19<01:58,  1.39s/it]
 87%|████████▋ | 549/633 [13:20<01:54,  1.36s/it]
 87%|████████▋ | 550/633 [13:21<01:48,  1.31s/it]
                                                 
{'loss': 0.2892, 'grad_norm': 2.0925703048706055, 'learning_rate': 2.795341098169717e-05, 'epoch': 1.3}

 87%|████████▋ | 550/633 [13:21<01:48,  1.31s/it]
 87%|████████▋ | 551/633 [13:23<01:48,  1.32s/it]
 87%|████████▋ | 552/633 [13:24<01:55,  1.42s/it]
 87%|████████▋ | 553/633 [13:26<02:00,  1.50s/it]
 88%|████████▊ | 554/633 [13:27<01:54,  1.45s/it]
 88%|████████▊ | 555/633 [13:29<01:51,  1.43s/it]
 88%|████████▊ | 556/633 [13:30<01:52,  1.46s/it]
 88%|████████▊ | 557/633 [13:32<01:46,  1.40s/it]
 88%|████████▊ | 558/633 [13:33<01:36,  1.29s/it]
 88%|████████▊ | 559/633 [13:34<01:44,  1.41s/it]
 88%|████████▊ | 560/633 [13:35<01:36,  1.32s/it]
                                                 
{'loss': 0.2369, 'grad_norm': 2.3992068767547607, 'learning_rate': 2.4625623960066556e-05, 'epoch': 1.33}

 88%|████████▊ | 560/633 [13:35<01:36,  1.32s/it]
 89%|████████▊ | 561/633 [13:37<01:38,  1.37s/it]
 89%|████████▉ | 562/633 [13:38<01:33,  1.32s/it]
 89%|████████▉ | 563/633 [13:39<01:23,  1.19s/it]
 89%|████████▉ | 564/633 [13:40<01:22,  1.20s/it]
 89%|████████▉ | 565/633 [13:42<01:31,  1.35s/it]
 89%|████████▉ | 566/633 [13:43<01:23,  1.25s/it]
 90%|████████▉ | 567/633 [13:44<01:18,  1.19s/it]
 90%|████████▉ | 568/633 [13:47<01:51,  1.71s/it]
 90%|████████▉ | 569/633 [13:48<01:46,  1.67s/it]
 90%|█████████ | 570/633 [13:50<01:37,  1.55s/it]
                                                 
{'loss': 0.2361, 'grad_norm': 1.664743423461914, 'learning_rate': 2.129783693843594e-05, 'epoch': 1.35}

 90%|█████████ | 570/633 [13:50<01:37,  1.55s/it]
 90%|█████████ | 571/633 [13:51<01:31,  1.48s/it]
 90%|█████████ | 572/633 [13:54<01:56,  1.91s/it]
 91%|█████████ | 573/633 [13:56<01:54,  1.91s/it]
 91%|█████████ | 574/633 [13:58<01:51,  1.88s/it]
 91%|█████████ | 575/633 [13:59<01:43,  1.79s/it]
 91%|█████████ | 576/633 [14:01<01:44,  1.83s/it]
 91%|█████████ | 577/633 [14:02<01:31,  1.64s/it]
 91%|█████████▏| 578/633 [14:04<01:26,  1.57s/it]
 91%|█████████▏| 579/633 [14:05<01:21,  1.51s/it]
 92%|█████████▏| 580/633 [14:06<01:13,  1.39s/it]
                                                 
{'loss': 0.2536, 'grad_norm': 1.4640089273452759, 'learning_rate': 1.7970049916805324e-05, 'epoch': 1.37}

 92%|█████████▏| 580/633 [14:06<01:13,  1.39s/it]
 92%|█████████▏| 581/633 [14:07<01:07,  1.31s/it]
 92%|█████████▏| 582/633 [14:08<01:04,  1.27s/it]
 92%|█████████▏| 583/633 [14:11<01:26,  1.72s/it]
 92%|█████████▏| 584/633 [14:12<01:14,  1.51s/it]
 92%|█████████▏| 585/633 [14:14<01:12,  1.52s/it]
 93%|█████████▎| 586/633 [14:15<01:08,  1.45s/it]
 93%|█████████▎| 587/633 [14:17<01:13,  1.60s/it]
 93%|█████████▎| 588/633 [14:18<01:02,  1.40s/it]
 93%|█████████▎| 589/633 [14:20<01:14,  1.70s/it]
 93%|█████████▎| 590/633 [14:22<01:13,  1.71s/it]
                                                 
{'loss': 0.239, 'grad_norm': 1.8241060972213745, 'learning_rate': 1.464226289517471e-05, 'epoch': 1.4}

 93%|█████████▎| 590/633 [14:22<01:13,  1.71s/it]
 93%|█████████▎| 591/633 [14:23<01:06,  1.57s/it]
 94%|█████████▎| 592/633 [14:25<01:00,  1.47s/it]
 94%|█████████▎| 593/633 [14:26<00:57,  1.44s/it]
 94%|█████████▍| 594/633 [14:27<00:55,  1.41s/it]
 94%|█████████▍| 595/633 [14:29<00:51,  1.36s/it]
 94%|█████████▍| 596/633 [14:30<00:49,  1.35s/it]
 94%|█████████▍| 597/633 [14:31<00:47,  1.33s/it]
 94%|█████████▍| 598/633 [14:32<00:41,  1.18s/it]
 95%|█████████▍| 599/633 [14:33<00:41,  1.21s/it]
 95%|█████████▍| 600/633 [14:35<00:40,  1.24s/it]
                                                 
{'loss': 0.2289, 'grad_norm': 3.208674192428589, 'learning_rate': 1.1314475873544094e-05, 'epoch': 1.42}

 95%|█████████▍| 600/633 [14:35<00:40,  1.24s/it]
 95%|█████████▍| 601/633 [14:37<00:50,  1.58s/it]
 95%|█████████▌| 602/633 [14:38<00:44,  1.44s/it]
 95%|█████████▌| 603/633 [14:39<00:41,  1.40s/it]
 95%|█████████▌| 604/633 [14:41<00:38,  1.32s/it]
 96%|█████████▌| 605/633 [14:42<00:42,  1.51s/it]
 96%|█████████▌| 606/633 [14:44<00:38,  1.43s/it]
 96%|█████████▌| 607/633 [14:45<00:39,  1.51s/it]
 96%|█████████▌| 608/633 [14:47<00:40,  1.63s/it]
 96%|█████████▌| 609/633 [14:48<00:34,  1.44s/it]
 96%|█████████▋| 610/633 [14:50<00:31,  1.37s/it]
                                                 
{'loss': 0.2486, 'grad_norm': 1.3943027257919312, 'learning_rate': 7.986688851913478e-06, 'epoch': 1.45}

 96%|█████████▋| 610/633 [14:50<00:31,  1.37s/it]
 97%|█████████▋| 611/633 [14:51<00:29,  1.35s/it]
 97%|█████████▋| 612/633 [14:52<00:28,  1.36s/it]
 97%|█████████▋| 613/633 [14:53<00:26,  1.33s/it]
 97%|█████████▋| 614/633 [14:55<00:24,  1.29s/it]
 97%|█████████▋| 615/633 [14:56<00:23,  1.33s/it]
 97%|█████████▋| 616/633 [14:58<00:23,  1.37s/it]
 97%|█████████▋| 617/633 [14:59<00:22,  1.38s/it]
 98%|█████████▊| 618/633 [15:00<00:20,  1.34s/it]
 98%|█████████▊| 619/633 [15:02<00:18,  1.35s/it]
 98%|█████████▊| 620/633 [15:03<00:18,  1.39s/it]
                                                 
{'loss': 0.2773, 'grad_norm': 1.491068959236145, 'learning_rate': 4.658901830282862e-06, 'epoch': 1.47}

 98%|█████████▊| 620/633 [15:03<00:18,  1.39s/it]
 98%|█████████▊| 621/633 [15:04<00:15,  1.25s/it]
 98%|█████████▊| 622/633 [15:06<00:15,  1.37s/it]
 98%|█████████▊| 623/633 [15:07<00:12,  1.29s/it]
 99%|█████████▊| 624/633 [15:08<00:10,  1.21s/it]
 99%|█████████▊| 625/633 [15:09<00:09,  1.21s/it]
 99%|█████████▉| 626/633 [15:10<00:08,  1.22s/it]
 99%|█████████▉| 627/633 [15:11<00:06,  1.09s/it]
 99%|█████████▉| 628/633 [15:12<00:05,  1.19s/it]
 99%|█████████▉| 629/633 [15:14<00:04,  1.16s/it]
100%|█████████▉| 630/633 [15:15<00:03,  1.16s/it]
                                                 
{'loss': 0.2665, 'grad_norm': 2.204336166381836, 'learning_rate': 1.3311148086522463e-06, 'epoch': 1.49}

100%|█████████▉| 630/633 [15:15<00:03,  1.16s/it]
100%|█████████▉| 631/633 [15:16<00:02,  1.16s/it]
100%|█████████▉| 632/633 [15:17<00:01,  1.22s/it]Saved adapter checkpoint checkpoints/modernbert-large-30k/checkpoint-4of4-step633

100%|██████████| 633/633 [15:19<00:00,  1.36s/it]

  0%|          | 0/47 [00:00<?, ?it/s][A

  4%|▍         | 2/47 [00:00<00:07,  5.98it/s][A

  6%|▋         | 3/47 [00:00<00:11,  3.96it/s][A

  9%|▊         | 4/47 [00:01<00:13,  3.29it/s][A

 11%|█         | 5/47 [00:01<00:13,  3.05it/s][A

 13%|█▎        | 6/47 [00:01<00:13,  3.05it/s][A

 15%|█▍        | 7/47 [00:02<00:13,  2.96it/s][A

 17%|█▋        | 8/47 [00:02<00:16,  2.41it/s][A

 19%|█▉        | 9/47 [00:03<00:15,  2.42it/s][A

 21%|██▏       | 10/47 [00:03<00:13,  2.66it/s][A

 23%|██▎       | 11/47 [00:03<00:13,  2.69it/s][A

 26%|██▌       | 12/47 [00:04<00:12,  2.75it/s][A

 28%|██▊       | 13/47 [00:04<00:12,  2.66it/s][A

 30%|██▉       | 14/47 [00:04<00:12,  2.73it/s][A

 32%|███▏      | 15/47 [00:05<00:11,  2.74it/s][A

 34%|███▍      | 16/47 [00:05<00:10,  2.83it/s][A

 36%|███▌      | 17/47 [00:05<00:10,  2.86it/s][A

 38%|███▊      | 18/47 [00:06<00:10,  2.75it/s][A

 40%|████      | 19/47 [00:06<00:09,  2.90it/s][A

 43%|████▎     | 20/47 [00:06<00:08,  3.03it/s][A

 45%|████▍     | 21/47 [00:07<00:10,  2.48it/s][A

 47%|████▋     | 22/47 [00:07<00:09,  2.53it/s][A

 49%|████▉     | 23/47 [00:08<00:08,  2.96it/s][A

 51%|█████     | 24/47 [00:08<00:08,  2.84it/s][A

 53%|█████▎    | 25/47 [00:08<00:07,  2.80it/s][A

 55%|█████▌    | 26/47 [00:09<00:07,  2.71it/s][A

 57%|█████▋    | 27/47 [00:09<00:07,  2.77it/s][A

 60%|█████▉    | 28/47 [00:09<00:06,  2.79it/s][A

 62%|██████▏   | 29/47 [00:10<00:06,  2.75it/s][A

 64%|██████▍   | 30/47 [00:10<00:06,  2.74it/s][A

 66%|██████▌   | 31/47 [00:11<00:05,  2.73it/s][A

 68%|██████▊   | 32/47 [00:11<00:05,  2.69it/s][A

 70%|███████   | 33/47 [00:11<00:05,  2.72it/s][A

 72%|███████▏  | 34/47 [00:12<00:04,  2.94it/s][A

 74%|███████▍  | 35/47 [00:12<00:04,  2.87it/s][A

 77%|███████▋  | 36/47 [00:12<00:03,  2.79it/s][A

 79%|███████▊  | 37/47 [00:13<00:03,  2.73it/s][A

 81%|████████  | 38/47 [00:13<00:03,  2.70it/s][A

 83%|████████▎ | 39/47 [00:14<00:03,  2.32it/s][A

 85%|████████▌ | 40/47 [00:14<00:02,  2.53it/s][A

 87%|████████▋ | 41/47 [00:15<00:02,  2.25it/s][A

 89%|████████▉ | 42/47 [00:15<00:02,  2.27it/s][A

 91%|█████████▏| 43/47 [00:15<00:01,  2.52it/s][A

 94%|█████████▎| 44/47 [00:16<00:01,  2.74it/s][A

 96%|█████████▌| 45/47 [00:16<00:00,  2.60it/s][A

 98%|█████████▊| 46/47 [00:16<00:00,  2.94it/s][A

100%|██████████| 47/47 [00:17<00:00,  2.87it/s][A
                                                 

                                               
[A{'eval_loss': 0.5154390931129456, 'eval_followed_precision': 0.8185880250223414, 'eval_followed_recall': 0.8215246636771301, 'eval_followed_f1': 0.8200537153088631, 'eval_followed_support': 1115, 'eval_violated_precision': 0.8070652173913043, 'eval_violated_recall': 0.7714285714285715, 'eval_violated_f1': 0.7888446215139442, 'eval_violated_support': 770, 'eval_not_applicable_precision': 0.8524017467248908, 'eval_not_applicable_recall': 0.8753363228699551, 'eval_not_applicable_f1': 0.863716814159292, 'eval_not_applicable_support': 1115, 'eval_accuracy': 0.8286666666666667, 'eval_macro_f1': 0.8242050503273665, 'eval_runtime': 17.87, 'eval_samples_per_second': 167.879, 'eval_steps_per_second': 2.63, 'epoch': 1.5}

100%|██████████| 633/633 [15:37<00:00,  1.36s/it]

100%|██████████| 47/47 [00:17<00:00,  2.87it/s][A

                                               [A
                                                 
{'train_runtime': 937.3102, 'train_samples_per_second': 43.209, 'train_steps_per_second': 0.675, 'train_loss': 0.4367924646178693, 'epoch': 1.5}

100%|██████████| 633/633 [15:37<00:00,  1.36s/it]
100%|██████████| 633/633 [15:37<00:00,  1.48s/it]

  0%|          | 0/47 [00:00<?, ?it/s]
  4%|▍         | 2/47 [00:00<00:07,  5.95it/s]
  6%|▋         | 3/47 [00:00<00:11,  3.96it/s]
  9%|▊         | 4/47 [00:01<00:13,  3.29it/s]
 11%|█         | 5/47 [00:01<00:13,  3.02it/s]
 13%|█▎        | 6/47 [00:01<00:13,  3.05it/s]
 15%|█▍        | 7/47 [00:02<00:13,  2.92it/s]
 17%|█▋        | 8/47 [00:02<00:16,  2.32it/s]
 19%|█▉        | 9/47 [00:03<00:15,  2.40it/s]
 21%|██▏       | 10/47 [00:03<00:13,  2.66it/s]
 23%|██▎       | 11/47 [00:03<00:13,  2.67it/s]
 26%|██▌       | 12/47 [00:04<00:12,  2.76it/s]
 28%|██▊       | 13/47 [00:04<00:12,  2.64it/s]
 30%|██▉       | 14/47 [00:04<00:12,  2.72it/s]
 32%|███▏      | 15/47 [00:05<00:11,  2.76it/s]
 34%|███▍      | 16/47 [00:05<00:10,  2.83it/s]
 36%|███▌      | 17/47 [00:05<00:10,  2.85it/s]
 38%|███▊      | 18/47 [00:06<00:10,  2.76it/s]
 40%|████      | 19/47 [00:06<00:09,  2.90it/s]
 43%|████▎     | 20/47 [00:06<00:08,  3.03it/s]
 45%|████▍     | 21/47 [00:07<00:10,  2.52it/s]
 47%|████▋     | 22/47 [00:07<00:09,  2.57it/s]
 49%|████▉     | 23/47 [00:08<00:07,  3.00it/s]
 51%|█████     | 24/47 [00:08<00:08,  2.87it/s]
 53%|█████▎    | 25/47 [00:08<00:07,  2.80it/s]
 55%|█████▌    | 26/47 [00:09<00:07,  2.73it/s]
 57%|█████▋    | 27/47 [00:09<00:07,  2.79it/s]
 60%|█████▉    | 28/47 [00:09<00:06,  2.80it/s]
 62%|██████▏   | 29/47 [00:10<00:06,  2.76it/s]
 64%|██████▍   | 30/47 [00:10<00:06,  2.75it/s]
 66%|██████▌   | 31/47 [00:11<00:05,  2.72it/s]
 68%|██████▊   | 32/47 [00:11<00:05,  2.70it/s]
 70%|███████   | 33/47 [00:11<00:05,  2.73it/s]
 72%|███████▏  | 34/47 [00:12<00:04,  2.95it/s]
 74%|███████▍  | 35/47 [00:12<00:04,  2.89it/s]
 77%|███████▋  | 36/47 [00:12<00:03,  2.80it/s]
 79%|███████▊  | 37/47 [00:13<00:03,  2.74it/s]
 81%|████████  | 38/47 [00:13<00:03,  2.69it/s]
 83%|████████▎ | 39/47 [00:14<00:03,  2.34it/s]
 85%|████████▌ | 40/47 [00:14<00:02,  2.52it/s]
 87%|████████▋ | 41/47 [00:15<00:02,  2.23it/s]
 89%|████████▉ | 42/47 [00:15<00:02,  2.25it/s]
 91%|█████████▏| 43/47 [00:15<00:01,  2.51it/s]
 94%|█████████▎| 44/47 [00:16<00:01,  2.72it/s]
 96%|█████████▌| 45/47 [00:16<00:00,  2.59it/s]
 98%|█████████▊| 46/47 [00:16<00:00,  2.93it/s]
100%|██████████| 47/47 [00:17<00:00,  2.87it/s]
100%|██████████| 47/47 [00:17<00:00,  2.75it/s]
                precision    recall  f1-score   support

      followed     0.8186    0.8215    0.8201      1115
      violated     0.8071    0.7714    0.7888       770
not_applicable     0.8524    0.8753    0.8637      1115

      accuracy                         0.8287      3000
     macro avg     0.8260    0.8228    0.8242      3000
  weighted avg     0.8282    0.8287    0.8283      3000

***** validation metrics *****
  validation_accuracy                 =     0.8287
  validation_followed_f1              =     0.8201
  validation_followed_precision       =     0.8186
  validation_followed_recall          =     0.8215
  validation_followed_support         =       1115
  validation_loss                     =     0.5154
  validation_macro_f1                 =     0.8242
  validation_not_applicable_f1        =     0.8637
  validation_not_applicable_precision =     0.8524
  validation_not_applicable_recall    =     0.8753
  validation_not_applicable_support   =       1115
  validation_runtime                  = 0:00:17.91
  validation_samples_per_second       =    167.458
  validation_steps_per_second         =      2.624
  validation_violated_f1              =     0.7888
  validation_violated_precision       =     0.8071
  validation_violated_recall          =     0.7714
  validation_violated_support         =        770
wandb: 
wandb: Run history:
wandb:                 eval/accuracy ▁▄▅█
wandb:              eval/followed_f1 ▁█▅█
wandb:       eval/followed_precision ▁▂██
wandb:          eval/followed_recall ▅█▁▂
wandb:         eval/followed_support ▁▁▁▁
wandb:                     eval/loss █▁▄▃
wandb:                 eval/macro_f1 ▁▅▆█
wandb:        eval/not_applicable_f1 ▇▁▇█
wandb: eval/not_applicable_precision ▁█▅▂
wandb:    eval/not_applicable_recall █▁▅█
wandb:                           +52 ...
wandb: 
wandb: Run summary:
wandb:                 eval/accuracy 0.82867
wandb:              eval/followed_f1 0.82005
wandb:       eval/followed_precision 0.81859
wandb:          eval/followed_recall 0.82152
wandb:         eval/followed_support 1115
wandb:                     eval/loss 0.51544
wandb:                 eval/macro_f1 0.82421
wandb:        eval/not_applicable_f1 0.86372
wandb: eval/not_applicable_precision 0.8524
wandb:    eval/not_applicable_recall 0.87534
wandb:                           +58 ...
wandb: 
wandb: You can sync this run to the cloud by running:
wandb: wandb sync /workspace/wandb/offline-run-20261103_143023-clywa1fj
wandb: Find logs at: ./wandb/offline-run-20261103_143023-clywa1fj/logs
Xet Storage Details

Size:: 78 kB
Xet hash:: aac20889432b1e8aa317971d440271e5a38b9da7876b6cbda6679cf2749cc31f
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.