============================================================
FLUX.2 Dev TE Fine-Tuning (LoRA)
============================================================

Loading tokenizer...

Loading TE model...
Loading safetensors weights...
Text model keys: 272
Extra keys (vision/tekken/etc): 223
Building MistralForCausalLM...
Tied keys (expected): ['lm_head.weight']
Model loaded. Parameters: 17,343,759,360

Applying LoRA...
trainable params: 138,608,640 || all params: 17,482,368,000 || trainable%: 0.7928

Loading captions...
Loaded 13956 captions

Map:   0%|          | 0/13956 [00:00<?, ? examples/s]
Map:   7%|▋         | 1000/13956 [00:00<00:04, 2757.55 examples/s]
Map:  14%|█▍        | 2000/13956 [00:00<00:04, 2981.19 examples/s]
Map:  21%|██▏       | 3000/13956 [00:00<00:03, 3062.14 examples/s]
Map:  29%|██▊       | 4000/13956 [00:01<00:03, 3066.22 examples/s]
Map:  36%|███▌      | 5000/13956 [00:01<00:02, 3086.92 examples/s]
Map:  43%|████▎     | 6000/13956 [00:01<00:02, 3077.69 examples/s]
Map:  50%|█████     | 7000/13956 [00:02<00:02, 3080.60 examples/s]
Map:  57%|█████▋    | 8000/13956 [00:02<00:01, 3093.12 examples/s]
Map:  64%|██████▍   | 9000/13956 [00:02<00:01, 3082.11 examples/s]
Map:  72%|███████▏  | 10000/13956 [00:03<00:01, 3072.51 examples/s]
Map:  79%|███████▉  | 11000/13956 [00:03<00:00, 3084.86 examples/s]
Map:  86%|████████▌ | 12000/13956 [00:04<00:00, 2594.51 examples/s]
Map:  93%|█████████▎| 13000/13956 [00:04<00:00, 2726.25 examples/s]
Map: 100%|██████████| 13956/13956 [00:04<00:00, 2826.87 examples/s]
Map: 100%|██████████| 13956/13956 [00:04<00:00, 2938.59 examples/s]
Train: 13676, Eval: 280
No label_names provided for model class `PeftModelForCausalLM`. Since `PeftModel` hides base models input arguments, if label_names is not given, label_names can't be set automatically within `Trainer`. Note that empty label_names list will be used instead.

Starting training...

  0%|          | 0/2562 [00:00<?, ?it/s]`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.

  0%|          | 1/2562 [00:04<3:03:20,  4.30s/it]
  0%|          | 2/2562 [00:07<2:25:09,  3.40s/it]
  0%|          | 3/2562 [00:10<2:19:04,  3.26s/it]
  0%|          | 4/2562 [00:13<2:12:52,  3.12s/it]
  0%|          | 5/2562 [00:16<2:10:16,  3.06s/it]
  0%|          | 6/2562 [00:19<2:09:29,  3.04s/it]
  0%|          | 7/2562 [00:21<2:08:01,  3.01s/it]
  0%|          | 8/2562 [00:24<2:06:08,  2.96s/it]
  0%|          | 9/2562 [00:28<2:09:31,  3.04s/it]
  0%|          | 10/2562 [00:30<2:07:45,  3.00s/it]
                                                   
{'loss': 12.0037, 'grad_norm': 2.8521690368652344, 'learning_rate': 1.550387596899225e-06, 'epoch': 0.01}

  0%|          | 10/2562 [00:30<2:07:45,  3.00s/it]
  0%|          | 11/2562 [00:33<2:05:14,  2.95s/it]
  0%|          | 12/2562 [00:36<2:05:11,  2.95s/it]
  1%|          | 13/2562 [00:39<2:02:38,  2.89s/it]
  1%|          | 14/2562 [00:42<2:03:41,  2.91s/it]
  1%|          | 15/2562 [00:45<2:03:41,  2.91s/it]
  1%|          | 16/2562 [00:48<2:02:45,  2.89s/it]
  1%|          | 17/2562 [00:50<2:01:12,  2.86s/it]
  1%|          | 18/2562 [00:54<2:04:16,  2.93s/it]
  1%|          | 19/2562 [00:57<2:04:07,  2.93s/it]
  1%|          | 20/2562 [00:59<2:02:19,  2.89s/it]
                                                   
{'loss': 11.9641, 'grad_norm': 3.560389280319214, 'learning_rate': 3.10077519379845e-06, 'epoch': 0.02}

  1%|          | 20/2562 [00:59<2:02:19,  2.89s/it]
  1%|          | 21/2562 [01:02<2:02:02,  2.88s/it]
  1%|          | 22/2562 [01:05<2:03:31,  2.92s/it]
  1%|          | 23/2562 [01:08<2:01:20,  2.87s/it]
  1%|          | 24/2562 [01:11<2:00:58,  2.86s/it]
  1%|          | 25/2562 [01:14<2:01:29,  2.87s/it]
  1%|          | 26/2562 [01:17<2:01:35,  2.88s/it]
  1%|          | 27/2562 [01:20<2:02:36,  2.90s/it]
  1%|          | 28/2562 [01:22<2:00:39,  2.86s/it]
  1%|          | 29/2562 [01:25<1:59:59,  2.84s/it]
  1%|          | 30/2562 [01:28<2:01:41,  2.88s/it]
                                                   
{'loss': 11.6641, 'grad_norm': 4.175055503845215, 'learning_rate': 4.651162790697675e-06, 'epoch': 0.04}

  1%|          | 30/2562 [01:28<2:01:41,  2.88s/it]
  1%|          | 31/2562 [01:31<2:00:45,  2.86s/it]
  1%|          | 32/2562 [01:34<2:02:31,  2.91s/it]
  1%|▏         | 33/2562 [01:37<2:02:42,  2.91s/it]
  1%|▏         | 34/2562 [01:40<2:00:34,  2.86s/it]
  1%|▏         | 35/2562 [01:42<1:59:26,  2.84s/it]
  1%|▏         | 36/2562 [01:45<2:00:45,  2.87s/it]
  1%|▏         | 37/2562 [01:48<2:00:15,  2.86s/it]
  1%|▏         | 38/2562 [01:51<1:59:54,  2.85s/it]
  2%|▏         | 39/2562 [01:54<2:03:12,  2.93s/it]
  2%|▏         | 40/2562 [01:57<2:00:30,  2.87s/it]
                                                   
{'loss': 10.9347, 'grad_norm': 7.465065956115723, 'learning_rate': 6.2015503875969e-06, 'epoch': 0.05}

  2%|▏         | 40/2562 [01:57<2:00:30,  2.87s/it]
  2%|▏         | 41/2562 [02:00<2:02:05,  2.91s/it]
  2%|▏         | 42/2562 [02:03<2:01:53,  2.90s/it]
  2%|▏         | 43/2562 [02:06<2:01:17,  2.89s/it]
  2%|▏         | 44/2562 [02:08<2:01:39,  2.90s/it]
  2%|▏         | 45/2562 [02:11<2:02:25,  2.92s/it]
  2%|▏         | 46/2562 [02:14<2:00:47,  2.88s/it]
  2%|▏         | 47/2562 [02:17<2:02:53,  2.93s/it]
  2%|▏         | 48/2562 [02:20<2:00:10,  2.87s/it]
  2%|▏         | 49/2562 [02:23<1:59:30,  2.85s/it]
  2%|▏         | 50/2562 [02:26<2:01:00,  2.89s/it]
                                                   
{'loss': 9.1016, 'grad_norm': 11.800537109375, 'learning_rate': 7.751937984496126e-06, 'epoch': 0.06}

  2%|▏         | 50/2562 [02:26<2:01:00,  2.89s/it]
  2%|▏         | 51/2562 [02:29<2:00:03,  2.87s/it]
  2%|▏         | 52/2562 [02:31<1:59:42,  2.86s/it]
  2%|▏         | 53/2562 [02:35<2:02:42,  2.93s/it]
  2%|▏         | 54/2562 [02:37<2:03:01,  2.94s/it]
  2%|▏         | 55/2562 [02:40<2:01:19,  2.90s/it]
  2%|▏         | 56/2562 [02:43<2:03:36,  2.96s/it]
  2%|▏         | 57/2562 [02:46<2:02:50,  2.94s/it]
  2%|▏         | 58/2562 [02:49<2:02:56,  2.95s/it]
  2%|▏         | 59/2562 [02:52<2:01:55,  2.92s/it]
  2%|▏         | 60/2562 [02:55<2:02:17,  2.93s/it]
                                                   
{'loss': 6.7264, 'grad_norm': 7.239933490753174, 'learning_rate': 9.30232558139535e-06, 'epoch': 0.07}

  2%|▏         | 60/2562 [02:55<2:02:17,  2.93s/it]
  2%|▏         | 61/2562 [02:58<2:00:59,  2.90s/it]
  2%|▏         | 62/2562 [03:01<2:04:24,  2.99s/it]
  2%|▏         | 63/2562 [03:04<2:01:40,  2.92s/it]
  2%|▏         | 64/2562 [03:07<2:02:42,  2.95s/it]
  3%|▎         | 65/2562 [03:10<2:01:59,  2.93s/it]
  3%|▎         | 66/2562 [03:13<2:01:34,  2.92s/it]
  3%|▎         | 67/2562 [03:16<2:01:27,  2.92s/it]
  3%|▎         | 68/2562 [03:18<1:58:35,  2.85s/it]
  3%|▎         | 69/2562 [03:21<1:58:39,  2.86s/it]
  3%|▎         | 70/2562 [03:24<2:01:04,  2.92s/it]
                                                   
{'loss': 5.38, 'grad_norm': 4.705348491668701, 'learning_rate': 1.0852713178294573e-05, 'epoch': 0.08}

  3%|▎         | 70/2562 [03:24<2:01:04,  2.92s/it]
  3%|▎         | 71/2562 [03:27<2:00:21,  2.90s/it]
  3%|▎         | 72/2562 [03:30<1:59:35,  2.88s/it]
  3%|▎         | 73/2562 [03:33<2:01:49,  2.94s/it]
  3%|▎         | 74/2562 [03:36<1:59:47,  2.89s/it]
  3%|▎         | 75/2562 [03:39<1:59:09,  2.87s/it]
  3%|▎         | 76/2562 [03:42<2:01:03,  2.92s/it]
  3%|▎         | 77/2562 [03:44<1:59:20,  2.88s/it]
  3%|▎         | 78/2562 [03:47<1:58:55,  2.87s/it]
  3%|▎         | 79/2562 [03:50<2:00:21,  2.91s/it]
  3%|▎         | 80/2562 [03:53<2:00:23,  2.91s/it]
                                                   
{'loss': 4.296, 'grad_norm': 4.169477939605713, 'learning_rate': 1.24031007751938e-05, 'epoch': 0.09}

  3%|▎         | 80/2562 [03:53<2:00:23,  2.91s/it]
  3%|▎         | 81/2562 [03:56<1:58:49,  2.87s/it]
  3%|▎         | 82/2562 [03:59<2:01:36,  2.94s/it]
  3%|▎         | 83/2562 [04:02<2:00:43,  2.92s/it]
  3%|▎         | 84/2562 [04:05<2:04:07,  3.01s/it]
  3%|▎         | 85/2562 [04:08<2:01:39,  2.95s/it]
  3%|▎         | 86/2562 [04:11<1:59:51,  2.90s/it]
  3%|▎         | 87/2562 [04:14<2:01:23,  2.94s/it]
  3%|▎         | 88/2562 [04:17<2:00:23,  2.92s/it]
  3%|▎         | 89/2562 [04:19<1:59:07,  2.89s/it]
  4%|▎         | 90/2562 [04:22<1:59:04,  2.89s/it]
                                                   
{'loss': 3.4328, 'grad_norm': 4.548755168914795, 'learning_rate': 1.3953488372093025e-05, 'epoch': 0.11}

  4%|▎         | 90/2562 [04:22<1:59:04,  2.89s/it]
  4%|▎         | 91/2562 [04:25<1:57:24,  2.85s/it]
  4%|▎         | 92/2562 [04:28<1:57:20,  2.85s/it]
  4%|▎         | 93/2562 [04:31<2:00:32,  2.93s/it]
  4%|▎         | 94/2562 [04:34<1:58:46,  2.89s/it]
  4%|▎         | 95/2562 [04:37<1:58:04,  2.87s/it]
  4%|▎         | 96/2562 [04:40<2:00:09,  2.92s/it]
  4%|▍         | 97/2562 [04:43<1:59:03,  2.90s/it]
  4%|▍         | 98/2562 [04:45<1:58:23,  2.88s/it]
  4%|▍         | 99/2562 [04:48<1:59:54,  2.92s/it]
  4%|▍         | 100/2562 [04:51<1:59:23,  2.91s/it]
                                                    
{'loss': 2.7285, 'grad_norm': 4.822308540344238, 'learning_rate': 1.550387596899225e-05, 'epoch': 0.12}

  4%|▍         | 100/2562 [04:51<1:59:23,  2.91s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.86it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.80it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.24it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.91it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.57it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.75it/s]

 10%|█         | 14/140 [00:01<00:12,  9.81it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.84it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.75it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.67it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.60it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.54it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.63it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.49it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.18it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.23it/s]

 20%|██        | 28/140 [00:02<00:11,  9.54it/s]

 21%|██        | 29/140 [00:02<00:11,  9.40it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.09it/s]

 22%|██▏       | 31/140 [00:03<00:11,  9.09it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.30it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.75it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.86it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.88it/s]

 28%|██▊       | 39/140 [00:03<00:10,  9.99it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.98it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.98it/s]

 30%|███       | 42/140 [00:04<00:09,  9.98it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.97it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.93it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.66it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.32it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.28it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.18it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.26it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.46it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.49it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.46it/s]

 40%|████      | 56/140 [00:05<00:08,  9.75it/s]

 41%|████      | 57/140 [00:05<00:08,  9.54it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.42it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.53it/s]

 44%|████▎     | 61/140 [00:06<00:08,  9.77it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.81it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.64it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.57it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.55it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.53it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.74it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.79it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.84it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.86it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.86it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.83it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.60it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.21it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.38it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.32it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.26it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.17it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.17it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.35it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.22it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.20it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.11it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.15it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.05it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.13it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.29it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.49it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.34it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.08it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.18it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.33it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.38it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.42it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.45it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.72it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.75it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.64it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.54it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.60it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.66it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.67it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.82it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.86it/s]

 82%|████████▏ | 115/140 [00:11<00:02,  9.58it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.48it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.60it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.37it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.35it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.47it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.17it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.14it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.30it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.26it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.41it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.32it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.09it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.86it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.62it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.73it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.06it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.53it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.69it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.55it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.56it/s]

100%|██████████| 140/140 [00:14<00:00,  8.29it/s]
                                                    

                                                 
{'eval_loss': 2.58227276802063, 'eval_runtime': 15.0683, 'eval_samples_per_second': 18.582, 'eval_steps_per_second': 9.291, 'epoch': 0.12}

  4%|▍         | 100/2562 [05:06<1:59:23,  2.91s/it]

100%|██████████| 140/140 [00:14<00:00,  8.29it/s]

                                                 
  4%|▍         | 101/2562 [05:09<5:03:25,  7.40s/it]
  4%|▍         | 102/2562 [05:12<4:10:06,  6.10s/it]
  4%|▍         | 103/2562 [05:15<3:30:06,  5.13s/it]
  4%|▍         | 104/2562 [05:18<3:00:32,  4.41s/it]
  4%|▍         | 105/2562 [05:21<2:44:21,  4.01s/it]
  4%|▍         | 106/2562 [05:24<2:31:07,  3.69s/it]
  4%|▍         | 107/2562 [05:27<2:22:16,  3.48s/it]
  4%|▍         | 108/2562 [05:30<2:15:58,  3.32s/it]
  4%|▍         | 109/2562 [05:33<2:09:12,  3.16s/it]
  4%|▍         | 110/2562 [05:36<2:07:50,  3.13s/it]
                                                    
{'loss': 2.4108, 'grad_norm': 5.240631103515625, 'learning_rate': 1.7054263565891473e-05, 'epoch': 0.13}

  4%|▍         | 110/2562 [05:36<2:07:50,  3.13s/it]
  4%|▍         | 111/2562 [05:38<2:03:56,  3.03s/it]
  4%|▍         | 112/2562 [05:41<2:01:57,  2.99s/it]
  4%|▍         | 113/2562 [05:44<2:00:56,  2.96s/it]
  4%|▍         | 114/2562 [05:47<2:00:25,  2.95s/it]
  4%|▍         | 115/2562 [05:50<1:58:54,  2.92s/it]
  5%|▍         | 116/2562 [05:53<1:59:09,  2.92s/it]
  5%|▍         | 117/2562 [05:56<1:59:19,  2.93s/it]
  5%|▍         | 118/2562 [05:59<1:58:45,  2.92s/it]
  5%|▍         | 119/2562 [06:02<1:59:37,  2.94s/it]
  5%|▍         | 120/2562 [06:05<1:58:06,  2.90s/it]
                                                    
{'loss': 2.1537, 'grad_norm': 5.79343318939209, 'learning_rate': 1.86046511627907e-05, 'epoch': 0.14}

  5%|▍         | 120/2562 [06:05<1:58:06,  2.90s/it]
  5%|▍         | 121/2562 [06:07<1:55:36,  2.84s/it]
  5%|▍         | 122/2562 [06:10<1:57:38,  2.89s/it]
  5%|▍         | 123/2562 [06:13<1:59:23,  2.94s/it]
  5%|▍         | 124/2562 [06:16<1:57:56,  2.90s/it]
  5%|▍         | 125/2562 [06:19<1:59:52,  2.95s/it]
  5%|▍         | 126/2562 [06:22<1:58:21,  2.92s/it]
  5%|▍         | 127/2562 [06:25<1:56:43,  2.88s/it]
  5%|▍         | 128/2562 [06:28<1:57:50,  2.91s/it]
  5%|▌         | 129/2562 [06:31<1:55:28,  2.85s/it]
  5%|▌         | 130/2562 [06:33<1:53:44,  2.81s/it]
                                                    
{'loss': 1.9403, 'grad_norm': 5.286383152008057, 'learning_rate': 1.9999991663467044e-05, 'epoch': 0.15}

  5%|▌         | 130/2562 [06:33<1:53:44,  2.81s/it]
  5%|▌         | 131/2562 [06:36<1:56:20,  2.87s/it]
  5%|▌         | 132/2562 [06:39<1:57:21,  2.90s/it]
  5%|▌         | 133/2562 [06:42<1:59:01,  2.94s/it]
  5%|▌         | 134/2562 [06:45<1:56:19,  2.87s/it]
  5%|▌         | 135/2562 [06:48<1:55:56,  2.87s/it]
  5%|▌         | 136/2562 [06:51<1:56:24,  2.88s/it]
  5%|▌         | 137/2562 [06:54<1:56:13,  2.88s/it]
  5%|▌         | 138/2562 [06:56<1:54:03,  2.82s/it]
  5%|▌         | 139/2562 [06:59<1:55:37,  2.86s/it]
  5%|▌         | 140/2562 [07:02<1:57:10,  2.90s/it]
                                                    
{'loss': 1.9036, 'grad_norm': 5.613925457000732, 'learning_rate': 1.9998991296330317e-05, 'epoch': 0.16}

  5%|▌         | 140/2562 [07:02<1:57:10,  2.90s/it]
  6%|▌         | 141/2562 [07:05<1:56:08,  2.88s/it]
  6%|▌         | 142/2562 [07:08<1:56:28,  2.89s/it]
  6%|▌         | 143/2562 [07:11<1:55:37,  2.87s/it]
  6%|▌         | 144/2562 [07:14<1:54:52,  2.85s/it]
  6%|▌         | 145/2562 [07:17<1:55:27,  2.87s/it]
  6%|▌         | 146/2562 [07:19<1:56:05,  2.88s/it]
  6%|▌         | 147/2562 [07:22<1:56:01,  2.88s/it]
  6%|▌         | 148/2562 [07:25<1:55:44,  2.88s/it]
  6%|▌         | 149/2562 [07:28<1:54:32,  2.85s/it]
  6%|▌         | 150/2562 [07:31<1:54:38,  2.85s/it]
                                                    
{'loss': 1.7314, 'grad_norm': 5.931254863739014, 'learning_rate': 1.999632381371545e-05, 'epoch': 0.18}

  6%|▌         | 150/2562 [07:31<1:54:38,  2.85s/it]
  6%|▌         | 151/2562 [07:34<1:57:14,  2.92s/it]
  6%|▌         | 152/2562 [07:37<1:56:06,  2.89s/it]
  6%|▌         | 153/2562 [07:40<1:54:50,  2.86s/it]
  6%|▌         | 154/2562 [07:42<1:55:23,  2.88s/it]
  6%|▌         | 155/2562 [07:45<1:56:00,  2.89s/it]
  6%|▌         | 156/2562 [07:48<1:55:51,  2.89s/it]
  6%|▌         | 157/2562 [07:51<1:57:20,  2.93s/it]
  6%|▌         | 158/2562 [07:54<1:55:40,  2.89s/it]
  6%|▌         | 159/2562 [07:57<1:55:03,  2.87s/it]
  6%|▌         | 160/2562 [08:00<1:54:51,  2.87s/it]
                                                    
{'loss': 1.7044, 'grad_norm': 6.149502754211426, 'learning_rate': 1.9991989660367463e-05, 'epoch': 0.19}

  6%|▌         | 160/2562 [08:00<1:54:51,  2.87s/it]
  6%|▋         | 161/2562 [08:03<1:55:00,  2.87s/it]
  6%|▋         | 162/2562 [08:05<1:54:26,  2.86s/it]
  6%|▋         | 163/2562 [08:08<1:55:04,  2.88s/it]
  6%|▋         | 164/2562 [08:11<1:55:05,  2.88s/it]
  6%|▋         | 165/2562 [08:14<1:54:07,  2.86s/it]
  6%|▋         | 166/2562 [08:17<1:54:51,  2.88s/it]
  7%|▋         | 167/2562 [08:20<1:54:14,  2.86s/it]
  7%|▋         | 168/2562 [08:23<1:53:54,  2.85s/it]
  7%|▋         | 169/2562 [08:26<1:55:11,  2.89s/it]
  7%|▋         | 170/2562 [08:29<1:55:56,  2.91s/it]
                                                    
{'loss': 1.6107, 'grad_norm': 5.55122184753418, 'learning_rate': 1.998598955891266e-05, 'epoch': 0.2}

  7%|▋         | 170/2562 [08:29<1:55:56,  2.91s/it]
  7%|▋         | 171/2562 [08:31<1:55:08,  2.89s/it]
  7%|▋         | 172/2562 [08:34<1:56:07,  2.92s/it]
  7%|▋         | 173/2562 [08:37<1:58:09,  2.97s/it]
  7%|▋         | 174/2562 [08:40<1:56:25,  2.93s/it]
  7%|▋         | 175/2562 [08:43<1:57:34,  2.96s/it]
  7%|▋         | 176/2562 [08:46<1:56:50,  2.94s/it]
  7%|▋         | 177/2562 [08:49<1:55:40,  2.91s/it]
  7%|▋         | 178/2562 [08:52<1:56:25,  2.93s/it]
  7%|▋         | 179/2562 [08:55<1:56:17,  2.93s/it]
  7%|▋         | 180/2562 [08:58<1:56:37,  2.94s/it]
                                                    
{'loss': 1.6523, 'grad_norm': 5.595592498779297, 'learning_rate': 1.9978324509738147e-05, 'epoch': 0.21}

  7%|▋         | 180/2562 [08:58<1:56:37,  2.94s/it]
  7%|▋         | 181/2562 [09:01<1:57:20,  2.96s/it]
  7%|▋         | 182/2562 [09:04<1:56:21,  2.93s/it]
  7%|▋         | 183/2562 [09:07<1:55:31,  2.91s/it]
  7%|▋         | 184/2562 [09:10<1:56:47,  2.95s/it]
  7%|▋         | 185/2562 [09:13<1:55:48,  2.92s/it]
  7%|▋         | 186/2562 [09:16<1:57:01,  2.96s/it]
  7%|▋         | 187/2562 [09:18<1:54:48,  2.90s/it]
  7%|▋         | 188/2562 [09:21<1:53:15,  2.86s/it]
  7%|▋         | 189/2562 [09:24<1:53:06,  2.86s/it]
  7%|▋         | 190/2562 [09:27<1:52:46,  2.85s/it]
                                                    
{'loss': 1.4891, 'grad_norm': 5.843740940093994, 'learning_rate': 1.9968995790825048e-05, 'epoch': 0.22}

  7%|▋         | 190/2562 [09:27<1:52:46,  2.85s/it]
  7%|▋         | 191/2562 [09:30<1:52:43,  2.85s/it]
  7%|▋         | 192/2562 [09:33<1:53:04,  2.86s/it]
  8%|▊         | 193/2562 [09:35<1:52:46,  2.86s/it]
  8%|▊         | 194/2562 [09:38<1:53:10,  2.87s/it]
  8%|▊         | 195/2562 [09:41<1:52:51,  2.86s/it]
  8%|▊         | 196/2562 [09:44<1:55:46,  2.94s/it]
  8%|▊         | 197/2562 [09:47<1:54:11,  2.90s/it]
  8%|▊         | 198/2562 [09:50<1:55:51,  2.94s/it]
  8%|▊         | 199/2562 [09:53<1:54:23,  2.90s/it]
  8%|▊         | 200/2562 [09:56<1:52:34,  2.86s/it]
                                                    
{'loss': 1.5192, 'grad_norm': 5.863370895385742, 'learning_rate': 1.995800495753542e-05, 'epoch': 0.23}

  8%|▊         | 200/2562 [09:56<1:52:34,  2.86s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.73it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.78it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.25it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.92it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.58it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.76it/s]

 10%|█         | 14/140 [00:01<00:12,  9.81it/s]

 11%|█         | 15/140 [00:01<00:12,  9.84it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.83it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.86it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.67it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.57it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.52it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.46it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.57it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.43it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.13it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.18it/s]

 20%|██        | 28/140 [00:02<00:11,  9.51it/s]

 21%|██        | 29/140 [00:02<00:11,  9.36it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.07it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.08it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.29it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.74it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.82it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.85it/s]

 28%|██▊       | 39/140 [00:03<00:10,  9.94it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.94it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 30%|███       | 42/140 [00:04<00:09,  9.94it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.95it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.94it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.72it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.38it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.32it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.21it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.26it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.47it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.45it/s]

 40%|████      | 56/140 [00:05<00:08,  9.74it/s]

 41%|████      | 57/140 [00:05<00:08,  9.53it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.43it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.52it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.63it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.85it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.66it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.58it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.55it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.49it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.73it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.84it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.88it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.87it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.88it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.64it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.26it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.39it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.31it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.24it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.14it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.12it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.33it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.22it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.19it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.06it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.01it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.08it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.29it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.43it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.28it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.04it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.14it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.31it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.34it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.48it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.29it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.37it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.71it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.73it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.63it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.51it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.58it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.64it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.67it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.86it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.86it/s]

 82%|████████▏ | 115/140 [00:11<00:02,  9.55it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.46it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.57it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.33it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.30it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.43it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.13it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.10it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.27it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.21it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.39it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.28it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.07it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.84it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.59it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.70it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.02it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.27it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.65it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.69it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.51it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.54it/s]

100%|██████████| 140/140 [00:14<00:00,  8.17it/s]
                                                    

                                                 
{'eval_loss': 1.5672245025634766, 'eval_runtime': 15.0901, 'eval_samples_per_second': 18.555, 'eval_steps_per_second': 9.278, 'epoch': 0.23}

  8%|▊         | 200/2562 [10:11<1:52:34,  2.86s/it]

100%|██████████| 140/140 [00:14<00:00,  8.17it/s]

                                                 
  8%|▊         | 201/2562 [10:14<4:52:26,  7.43s/it]
  8%|▊         | 202/2562 [10:17<3:57:20,  6.03s/it]
  8%|▊         | 203/2562 [10:19<3:19:39,  5.08s/it]
  8%|▊         | 204/2562 [10:22<2:54:04,  4.43s/it]
  8%|▊         | 205/2562 [10:25<2:36:03,  3.97s/it]
  8%|▊         | 206/2562 [10:28<2:23:59,  3.67s/it]
  8%|▊         | 207/2562 [10:31<2:14:23,  3.42s/it]
  8%|▊         | 208/2562 [10:34<2:07:20,  3.25s/it]
  8%|▊         | 209/2562 [10:37<2:02:17,  3.12s/it]
  8%|▊         | 210/2562 [10:40<2:01:32,  3.10s/it]
                                                    
{'loss': 1.4783, 'grad_norm': 5.498380661010742, 'learning_rate': 1.9945353842352943e-05, 'epoch': 0.25}

  8%|▊         | 210/2562 [10:40<2:01:32,  3.10s/it]
  8%|▊         | 211/2562 [10:43<1:57:54,  3.01s/it]
  8%|▊         | 212/2562 [10:45<1:56:18,  2.97s/it]
  8%|▊         | 213/2562 [10:49<1:57:37,  3.00s/it]
  8%|▊         | 214/2562 [10:51<1:55:01,  2.94s/it]
  8%|▊         | 215/2562 [10:54<1:53:33,  2.90s/it]
  8%|▊         | 216/2562 [10:57<1:54:50,  2.94s/it]
  8%|▊         | 217/2562 [11:00<1:52:29,  2.88s/it]
  9%|▊         | 218/2562 [11:03<1:52:44,  2.89s/it]
  9%|▊         | 219/2562 [11:06<1:51:12,  2.85s/it]
  9%|▊         | 220/2562 [11:08<1:52:21,  2.88s/it]
                                                    
{'loss': 1.4733, 'grad_norm': 5.338771820068359, 'learning_rate': 1.9931044554577373e-05, 'epoch': 0.26}

  9%|▊         | 220/2562 [11:09<1:52:21,  2.88s/it]
  9%|▊         | 221/2562 [11:11<1:52:45,  2.89s/it]
  9%|▊         | 222/2562 [11:14<1:51:44,  2.87s/it]
  9%|▊         | 223/2562 [11:17<1:51:48,  2.87s/it]
  9%|▊         | 224/2562 [11:20<1:52:43,  2.89s/it]
  9%|▉         | 225/2562 [11:23<1:51:54,  2.87s/it]
  9%|▉         | 226/2562 [11:26<1:52:45,  2.90s/it]
  9%|▉         | 227/2562 [11:29<1:53:56,  2.93s/it]
  9%|▉         | 228/2562 [11:32<1:51:47,  2.87s/it]
  9%|▉         | 229/2562 [11:34<1:51:53,  2.88s/it]
  9%|▉         | 230/2562 [11:37<1:53:30,  2.92s/it]
                                                    
{'loss': 1.4471, 'grad_norm': 5.779300689697266, 'learning_rate': 1.991507947997287e-05, 'epoch': 0.27}

  9%|▉         | 230/2562 [11:37<1:53:30,  2.92s/it]
  9%|▉         | 231/2562 [11:40<1:53:23,  2.92s/it]
  9%|▉         | 232/2562 [11:43<1:52:12,  2.89s/it]
  9%|▉         | 233/2562 [11:46<1:54:25,  2.95s/it]
  9%|▉         | 234/2562 [11:49<1:52:43,  2.91s/it]
  9%|▉         | 235/2562 [11:52<1:53:16,  2.92s/it]
  9%|▉         | 236/2562 [11:55<1:54:38,  2.96s/it]
  9%|▉         | 237/2562 [11:58<1:53:51,  2.94s/it]
  9%|▉         | 238/2562 [12:01<1:52:08,  2.90s/it]
  9%|▉         | 239/2562 [12:04<1:53:58,  2.94s/it]
  9%|▉         | 240/2562 [12:07<1:52:01,  2.89s/it]
                                                    
{'loss': 1.4807, 'grad_norm': 6.205949783325195, 'learning_rate': 1.989746128037024e-05, 'epoch': 0.28}

  9%|▉         | 240/2562 [12:07<1:52:01,  2.89s/it]
  9%|▉         | 241/2562 [12:09<1:51:16,  2.88s/it]
  9%|▉         | 242/2562 [12:12<1:51:04,  2.87s/it]
  9%|▉         | 243/2562 [12:15<1:50:08,  2.85s/it]
 10%|▉         | 244/2562 [12:18<1:51:37,  2.89s/it]
 10%|▉         | 245/2562 [12:21<1:50:51,  2.87s/it]
 10%|▉         | 246/2562 [12:24<1:51:32,  2.89s/it]
 10%|▉         | 247/2562 [12:27<1:51:59,  2.90s/it]
 10%|▉         | 248/2562 [12:30<1:51:13,  2.88s/it]
 10%|▉         | 249/2562 [12:32<1:49:08,  2.83s/it]
 10%|▉         | 250/2562 [12:35<1:51:04,  2.88s/it]
                                                    
{'loss': 1.4084, 'grad_norm': 5.171810626983643, 'learning_rate': 1.987819289322311e-05, 'epoch': 0.29}

 10%|▉         | 250/2562 [12:35<1:51:04,  2.88s/it]
 10%|▉         | 251/2562 [12:38<1:50:12,  2.86s/it]
 10%|▉         | 252/2562 [12:41<1:51:10,  2.89s/it]
 10%|▉         | 253/2562 [12:44<1:51:52,  2.91s/it]
 10%|▉         | 254/2562 [12:47<1:52:34,  2.93s/it]
 10%|▉         | 255/2562 [12:50<1:51:41,  2.90s/it]
 10%|▉         | 256/2562 [12:53<1:52:28,  2.93s/it]
 10%|█         | 257/2562 [12:56<1:52:33,  2.93s/it]
 10%|█         | 258/2562 [12:59<1:50:35,  2.88s/it]
 10%|█         | 259/2562 [13:02<1:53:06,  2.95s/it]
 10%|█         | 260/2562 [13:05<1:51:52,  2.92s/it]
                                                    
{'loss': 1.4083, 'grad_norm': 5.690944671630859, 'learning_rate': 1.9857277531118173e-05, 'epoch': 0.3}

 10%|█         | 260/2562 [13:05<1:51:52,  2.92s/it]
 10%|█         | 261/2562 [13:08<1:52:37,  2.94s/it]
 10%|█         | 262/2562 [13:10<1:52:56,  2.95s/it]
 10%|█         | 263/2562 [13:13<1:52:35,  2.94s/it]
 10%|█         | 264/2562 [13:16<1:52:07,  2.93s/it]
 10%|█         | 265/2562 [13:19<1:51:11,  2.90s/it]
 10%|█         | 266/2562 [13:22<1:51:25,  2.91s/it]
 10%|█         | 267/2562 [13:25<1:52:35,  2.94s/it]
 10%|█         | 268/2562 [13:28<1:53:37,  2.97s/it]
 10%|█         | 269/2562 [13:31<1:53:28,  2.97s/it]
 11%|█         | 270/2562 [13:34<1:52:54,  2.96s/it]
                                                    
{'loss': 1.371, 'grad_norm': 5.631648540496826, 'learning_rate': 1.983471868123958e-05, 'epoch': 0.32}

 11%|█         | 270/2562 [13:34<1:52:54,  2.96s/it]
 11%|█         | 271/2562 [13:37<1:51:11,  2.91s/it]
 11%|█         | 272/2562 [13:40<1:50:01,  2.88s/it]
 11%|█         | 273/2562 [13:43<1:52:01,  2.94s/it]
 11%|█         | 274/2562 [13:45<1:49:16,  2.87s/it]
 11%|█         | 275/2562 [13:48<1:50:07,  2.89s/it]
 11%|█         | 276/2562 [13:51<1:50:20,  2.90s/it]
 11%|█         | 277/2562 [13:54<1:49:59,  2.89s/it]
 11%|█         | 278/2562 [13:57<1:48:40,  2.85s/it]
 11%|█         | 279/2562 [14:00<1:49:27,  2.88s/it]
 11%|█         | 280/2562 [14:03<1:49:30,  2.88s/it]
                                                    
{'loss': 1.3456, 'grad_norm': 5.213723659515381, 'learning_rate': 1.98105201047875e-05, 'epoch': 0.33}

 11%|█         | 280/2562 [14:03<1:49:30,  2.88s/it]
 11%|█         | 281/2562 [14:06<1:50:03,  2.90s/it]
 11%|█         | 282/2562 [14:09<1:52:15,  2.95s/it]
 11%|█         | 283/2562 [14:12<1:50:27,  2.91s/it]
 11%|█         | 284/2562 [14:14<1:50:07,  2.90s/it]
 11%|█         | 285/2562 [14:17<1:51:45,  2.94s/it]
 11%|█         | 286/2562 [14:21<1:52:42,  2.97s/it]
 11%|█         | 287/2562 [14:23<1:50:49,  2.92s/it]
 11%|█         | 288/2562 [14:26<1:53:30,  3.00s/it]
 11%|█▏        | 289/2562 [14:29<1:50:52,  2.93s/it]
 11%|█▏        | 290/2562 [14:32<1:51:11,  2.94s/it]
                                                    
{'loss': 1.3091, 'grad_norm': 5.488040924072266, 'learning_rate': 1.9784685836351045e-05, 'epoch': 0.34}

 11%|█▏        | 290/2562 [14:32<1:51:11,  2.94s/it]
 11%|█▏        | 291/2562 [14:35<1:51:36,  2.95s/it]
 11%|█▏        | 292/2562 [14:38<1:49:17,  2.89s/it]
 11%|█▏        | 293/2562 [14:41<1:49:09,  2.89s/it]
 11%|█▏        | 294/2562 [14:44<1:48:30,  2.87s/it]
 12%|█▏        | 295/2562 [14:47<1:48:43,  2.88s/it]
 12%|█▏        | 296/2562 [14:49<1:49:06,  2.89s/it]
 12%|█▏        | 297/2562 [14:52<1:50:06,  2.92s/it]
 12%|█▏        | 298/2562 [14:55<1:48:51,  2.89s/it]
 12%|█▏        | 299/2562 [14:58<1:49:48,  2.91s/it]
 12%|█▏        | 300/2562 [15:01<1:50:40,  2.94s/it]
                                                    
{'loss': 1.3599, 'grad_norm': 5.875193119049072, 'learning_rate': 1.975722018323556e-05, 'epoch': 0.35}

 12%|█▏        | 300/2562 [15:01<1:50:40,  2.94s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.57it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.70it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.16it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.84it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.52it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.76it/s]

 10%|█         | 14/140 [00:01<00:12,  9.77it/s]

 11%|█         | 15/140 [00:01<00:12,  9.79it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.77it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.80it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.58it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.53it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.48it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.42it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.52it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.40it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.10it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.16it/s]

 20%|██        | 28/140 [00:02<00:11,  9.51it/s]

 21%|██        | 29/140 [00:02<00:11,  9.35it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.05it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.05it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.28it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.75it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.84it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.85it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.95it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.95it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.91it/s]

 30%|███       | 42/140 [00:04<00:09,  9.90it/s]

 31%|███       | 43/140 [00:04<00:09,  9.87it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.85it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.83it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.56it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.21it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.22it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.13it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.21it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.40it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.45it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.46it/s]

 40%|████      | 56/140 [00:05<00:08,  9.73it/s]

 41%|████      | 57/140 [00:05<00:08,  9.50it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.40it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.50it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.62it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.83it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.64it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.58it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.57it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.55it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.75it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.81it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.84it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.85it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.85it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.87it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.59it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.17it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.34it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.30it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.22it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.12it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.12it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.27it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.18it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.17it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.07it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.98it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.06it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.27it/s]

 66%|██████▋   | 93/140 [00:09<00:05,  9.37it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.23it/s]

 68%|██████▊   | 95/140 [00:09<00:05,  9.00it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.11it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.29it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.35it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.50it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.32it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.37it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.53it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.74it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.62it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.52it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.58it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.62it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.64it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.81it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.81it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.51it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.40it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.51it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.29it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.26it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.39it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.10it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.09it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.23it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.34it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.15it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.93it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.68it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.77it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  9.05it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.28it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.65it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.69it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.52it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.53it/s]

100%|██████████| 140/140 [00:14<00:00,  8.17it/s]
                                                    

                                                 
{'eval_loss': 1.3748310804367065, 'eval_runtime': 15.1278, 'eval_samples_per_second': 18.509, 'eval_steps_per_second': 9.255, 'epoch': 0.35}

 12%|█▏        | 300/2562 [15:16<1:50:40,  2.94s/it]

100%|██████████| 140/140 [00:14<00:00,  8.17it/s]

                                                 
 12%|█▏        | 301/2562 [15:19<4:41:04,  7.46s/it]
 12%|█▏        | 302/2562 [15:22<3:51:28,  6.15s/it]
 12%|█▏        | 303/2562 [15:25<3:14:29,  5.17s/it]
 12%|█▏        | 304/2562 [15:28<2:50:14,  4.52s/it]
 12%|█▏        | 305/2562 [15:31<2:33:16,  4.07s/it]
 12%|█▏        | 306/2562 [15:34<2:21:23,  3.76s/it]
 12%|█▏        | 307/2562 [15:37<2:11:27,  3.50s/it]
 12%|█▏        | 308/2562 [15:40<2:04:57,  3.33s/it]
 12%|█▏        | 309/2562 [15:43<2:01:58,  3.25s/it]
 12%|█▏        | 310/2562 [15:46<1:57:33,  3.13s/it]
                                                    
{'loss': 1.3253, 'grad_norm': 6.283113956451416, 'learning_rate': 1.9728127724744516e-05, 'epoch': 0.36}

 12%|█▏        | 310/2562 [15:46<1:57:33,  3.13s/it]
 12%|█▏        | 311/2562 [15:49<1:58:18,  3.15s/it]
 12%|█▏        | 312/2562 [15:52<1:54:08,  3.04s/it]
 12%|█▏        | 313/2562 [15:55<1:52:37,  3.00s/it]
 12%|█▏        | 314/2562 [15:58<1:52:22,  3.00s/it]
 12%|█▏        | 315/2562 [16:01<1:49:36,  2.93s/it]
 12%|█▏        | 316/2562 [16:03<1:48:08,  2.89s/it]
 12%|█▏        | 317/2562 [16:06<1:48:37,  2.90s/it]
 12%|█▏        | 318/2562 [16:09<1:48:41,  2.91s/it]
 12%|█▏        | 319/2562 [16:12<1:46:24,  2.85s/it]
 12%|█▏        | 320/2562 [16:15<1:48:18,  2.90s/it]
                                                    
{'loss': 1.3722, 'grad_norm': 5.70245885848999, 'learning_rate': 1.9697413311415967e-05, 'epoch': 0.37}

 12%|█▏        | 320/2562 [16:15<1:48:18,  2.90s/it]
 13%|█▎        | 321/2562 [16:18<1:47:47,  2.89s/it]
 13%|█▎        | 322/2562 [16:21<1:49:33,  2.93s/it]
 13%|█▎        | 323/2562 [16:24<1:48:15,  2.90s/it]
 13%|█▎        | 324/2562 [16:27<1:47:43,  2.89s/it]
 13%|█▎        | 325/2562 [16:30<1:50:14,  2.96s/it]
 13%|█▎        | 326/2562 [16:33<1:48:33,  2.91s/it]
 13%|█▎        | 327/2562 [16:35<1:47:12,  2.88s/it]
 13%|█▎        | 328/2562 [16:38<1:47:54,  2.90s/it]
 13%|█▎        | 329/2562 [16:41<1:47:52,  2.90s/it]
 13%|█▎        | 330/2562 [16:44<1:45:37,  2.84s/it]
                                                    
{'loss': 1.2583, 'grad_norm': 5.350372791290283, 'learning_rate': 1.9665082064213856e-05, 'epoch': 0.39}

 13%|█▎        | 330/2562 [16:44<1:45:37,  2.84s/it]
 13%|█▎        | 331/2562 [16:47<1:46:24,  2.86s/it]
 13%|█▎        | 332/2562 [16:50<1:46:07,  2.86s/it]
 13%|█▎        | 333/2562 [16:52<1:44:53,  2.82s/it]
 13%|█▎        | 334/2562 [16:55<1:46:31,  2.87s/it]
 13%|█▎        | 335/2562 [16:58<1:45:12,  2.83s/it]
 13%|█▎        | 336/2562 [17:01<1:44:22,  2.81s/it]
 13%|█▎        | 337/2562 [17:04<1:48:11,  2.92s/it]
 13%|█▎        | 338/2562 [17:07<1:48:48,  2.94s/it]
 13%|█▎        | 339/2562 [17:10<1:49:17,  2.95s/it]
 13%|█▎        | 340/2562 [17:13<1:48:13,  2.92s/it]
                                                    
{'loss': 1.2911, 'grad_norm': 5.489436149597168, 'learning_rate': 1.9631139373674188e-05, 'epoch': 0.4}

 13%|█▎        | 340/2562 [17:13<1:48:13,  2.92s/it]
 13%|█▎        | 341/2562 [17:16<1:47:42,  2.91s/it]
 13%|█▎        | 342/2562 [17:19<1:47:48,  2.91s/it]
 13%|█▎        | 343/2562 [17:22<1:47:51,  2.92s/it]
 13%|█▎        | 344/2562 [17:25<1:48:29,  2.93s/it]
 13%|█▎        | 345/2562 [17:27<1:48:12,  2.93s/it]
 14%|█▎        | 346/2562 [17:30<1:47:28,  2.91s/it]
 14%|█▎        | 347/2562 [17:33<1:45:36,  2.86s/it]
 14%|█▎        | 348/2562 [17:36<1:48:56,  2.95s/it]
 14%|█▎        | 349/2562 [17:39<1:46:44,  2.89s/it]
 14%|█▎        | 350/2562 [17:42<1:45:22,  2.86s/it]
                                                    
{'loss': 1.3038, 'grad_norm': 5.308193683624268, 'learning_rate': 1.9595590899006288e-05, 'epoch': 0.41}

 14%|█▎        | 350/2562 [17:42<1:45:22,  2.86s/it]
 14%|█▎        | 351/2562 [17:45<1:48:47,  2.95s/it]
 14%|█▎        | 352/2562 [17:48<1:47:24,  2.92s/it]
 14%|█▍        | 353/2562 [17:51<1:45:29,  2.87s/it]
 14%|█▍        | 354/2562 [17:53<1:46:11,  2.89s/it]
 14%|█▍        | 355/2562 [17:56<1:44:41,  2.85s/it]
 14%|█▍        | 356/2562 [17:59<1:44:56,  2.85s/it]
 14%|█▍        | 357/2562 [18:02<1:46:16,  2.89s/it]
 14%|█▍        | 358/2562 [18:05<1:47:20,  2.92s/it]
 14%|█▍        | 359/2562 [18:08<1:47:09,  2.92s/it]
 14%|█▍        | 360/2562 [18:11<1:45:55,  2.89s/it]
                                                    
{'loss': 1.2607, 'grad_norm': 5.691999435424805, 'learning_rate': 1.9558442567149244e-05, 'epoch': 0.42}

 14%|█▍        | 360/2562 [18:11<1:45:55,  2.89s/it]
 14%|█▍        | 361/2562 [18:14<1:47:08,  2.92s/it]
 14%|█▍        | 362/2562 [18:17<1:49:26,  2.98s/it]
 14%|█▍        | 363/2562 [18:20<1:47:47,  2.94s/it]
 14%|█▍        | 364/2562 [18:23<1:46:13,  2.90s/it]
 14%|█▍        | 365/2562 [18:26<1:48:35,  2.97s/it]
 14%|█▍        | 366/2562 [18:29<1:46:57,  2.92s/it]
 14%|█▍        | 367/2562 [18:32<1:47:49,  2.95s/it]
 14%|█▍        | 368/2562 [18:35<1:51:00,  3.04s/it]
 14%|█▍        | 369/2562 [18:38<1:48:33,  2.97s/it]
 14%|█▍        | 370/2562 [18:40<1:47:28,  2.94s/it]
                                                    
{'loss': 1.3111, 'grad_norm': 5.67365026473999, 'learning_rate': 1.9519700571783718e-05, 'epoch': 0.43}

 14%|█▍        | 370/2562 [18:40<1:47:28,  2.94s/it]
 14%|█▍        | 371/2562 [18:43<1:47:01,  2.93s/it]
 15%|█▍        | 372/2562 [18:46<1:46:05,  2.91s/it]
 15%|█▍        | 373/2562 [18:49<1:44:15,  2.86s/it]
 15%|█▍        | 374/2562 [18:52<1:45:01,  2.88s/it]
 15%|█▍        | 375/2562 [18:55<1:43:44,  2.85s/it]
 15%|█▍        | 376/2562 [18:58<1:43:40,  2.85s/it]
 15%|█▍        | 377/2562 [19:00<1:45:05,  2.89s/it]
 15%|█▍        | 378/2562 [19:03<1:43:51,  2.85s/it]
 15%|█▍        | 379/2562 [19:06<1:43:44,  2.85s/it]
 15%|█▍        | 380/2562 [19:09<1:44:05,  2.86s/it]
                                                    
{'loss': 1.2663, 'grad_norm': 5.552055835723877, 'learning_rate': 1.947937137229928e-05, 'epoch': 0.44}

 15%|█▍        | 380/2562 [19:09<1:44:05,  2.86s/it]
 15%|█▍        | 381/2562 [19:12<1:44:23,  2.87s/it]
 15%|█▍        | 382/2562 [19:15<1:45:57,  2.92s/it]
 15%|█▍        | 383/2562 [19:18<1:44:01,  2.86s/it]
 15%|█▍        | 384/2562 [19:21<1:43:59,  2.86s/it]
 15%|█▌        | 385/2562 [19:24<1:46:17,  2.93s/it]
 15%|█▌        | 386/2562 [19:27<1:46:37,  2.94s/it]
 15%|█▌        | 387/2562 [19:30<1:47:04,  2.95s/it]
 15%|█▌        | 388/2562 [19:33<1:47:20,  2.96s/it]
 15%|█▌        | 389/2562 [19:35<1:46:32,  2.94s/it]
 15%|█▌        | 390/2562 [19:38<1:44:33,  2.89s/it]
                                                    
{'loss': 1.2883, 'grad_norm': 5.420053005218506, 'learning_rate': 1.943746169271746e-05, 'epoch': 0.46}

 15%|█▌        | 390/2562 [19:38<1:44:33,  2.89s/it]
 15%|█▌        | 391/2562 [19:41<1:45:54,  2.93s/it]
 15%|█▌        | 392/2562 [19:44<1:44:41,  2.89s/it]
 15%|█▌        | 393/2562 [19:47<1:46:31,  2.95s/it]
 15%|█▌        | 394/2562 [19:50<1:47:29,  2.97s/it]
 15%|█▌        | 395/2562 [19:53<1:44:30,  2.89s/it]
 15%|█▌        | 396/2562 [19:56<1:45:59,  2.94s/it]
 15%|█▌        | 397/2562 [19:59<1:44:47,  2.90s/it]
 16%|█▌        | 398/2562 [20:02<1:44:31,  2.90s/it]
 16%|█▌        | 399/2562 [20:04<1:43:21,  2.87s/it]
 16%|█▌        | 400/2562 [20:07<1:45:14,  2.92s/it]
                                                    
{'loss': 1.2424, 'grad_norm': 5.484871864318848, 'learning_rate': 1.9393978520570638e-05, 'epoch': 0.47}

 16%|█▌        | 400/2562 [20:07<1:45:14,  2.92s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.60it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.67it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.19it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.87it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.54it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.78it/s]

 11%|█         | 15/140 [00:01<00:12,  9.81it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.80it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.82it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.59it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.53it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.48it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.43it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.54it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.39it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.09it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.15it/s]

 20%|██        | 28/140 [00:02<00:11,  9.46it/s]

 21%|██        | 29/140 [00:02<00:11,  9.30it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.02it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.01it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.21it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.67it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.78it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.80it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.91it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.92it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.90it/s]

 30%|███       | 42/140 [00:04<00:09,  9.91it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.91it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.88it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.66it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.33it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.25it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.17it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.22it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.40it/s]

 37%|███▋      | 52/140 [00:05<00:09,  9.56it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.33it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.38it/s]

 40%|████      | 56/140 [00:05<00:08,  9.71it/s]

 41%|████      | 57/140 [00:05<00:08,  9.48it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.39it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.61it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.83it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.64it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.56it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.53it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.49it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.70it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.76it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.81it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.85it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.87it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.86it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.87it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.60it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.21it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.36it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.28it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.20it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.09it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.12it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.32it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.20it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.20it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.11it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.12it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.98it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.07it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.27it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.44it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.27it/s]

 68%|██████▊   | 95/140 [00:09<00:05,  9.00it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.11it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.36it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.23it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.26it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.45it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.25it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.32it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.49it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.72it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.61it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.51it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.60it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.64it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.66it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.83it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.84it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.53it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.55it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.32it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.30it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.12it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.08it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.22it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.37it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.27it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.04it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.83it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.59it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.71it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  9.02it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.27it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.58it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.67it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.49it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.51it/s]

100%|██████████| 140/140 [00:14<00:00,  8.06it/s]
                                                    

                                                 
{'eval_loss': 1.2776166200637817, 'eval_runtime': 15.1406, 'eval_samples_per_second': 18.493, 'eval_steps_per_second': 9.247, 'epoch': 0.47}

 16%|█▌        | 400/2562 [20:23<1:45:14,  2.92s/it]

100%|██████████| 140/140 [00:14<00:00,  8.06it/s]

                                                 
 16%|█▌        | 401/2562 [20:25<4:28:10,  7.45s/it]
 16%|█▌        | 402/2562 [20:28<3:39:41,  6.10s/it]
 16%|█▌        | 403/2562 [20:31<3:04:30,  5.13s/it]
 16%|█▌        | 404/2562 [20:34<2:39:39,  4.44s/it]
 16%|█▌        | 405/2562 [20:37<2:23:47,  4.00s/it]
 16%|█▌        | 406/2562 [20:40<2:11:41,  3.67s/it]
 16%|█▌        | 407/2562 [20:43<2:03:02,  3.43s/it]
 16%|█▌        | 408/2562 [20:46<1:58:07,  3.29s/it]
 16%|█▌        | 409/2562 [20:49<1:52:53,  3.15s/it]
 16%|█▌        | 410/2562 [20:51<1:48:00,  3.01s/it]
                                                    
{'loss': 1.2303, 'grad_norm': 5.9503350257873535, 'learning_rate': 1.9348929105737044e-05, 'epoch': 0.48}

 16%|█▌        | 410/2562 [20:51<1:48:00,  3.01s/it]
 16%|█▌        | 411/2562 [20:54<1:47:20,  2.99s/it]
 16%|█▌        | 412/2562 [20:57<1:45:41,  2.95s/it]
 16%|█▌        | 413/2562 [21:00<1:44:55,  2.93s/it]
 16%|█▌        | 414/2562 [21:03<1:45:43,  2.95s/it]
 16%|█▌        | 415/2562 [21:06<1:44:47,  2.93s/it]
 16%|█▌        | 416/2562 [21:09<1:44:37,  2.93s/it]
 16%|█▋        | 417/2562 [21:12<1:45:44,  2.96s/it]
 16%|█▋        | 418/2562 [21:15<1:44:26,  2.92s/it]
 16%|█▋        | 419/2562 [21:17<1:42:29,  2.87s/it]
 16%|█▋        | 420/2562 [21:20<1:43:34,  2.90s/it]
                                                    
{'loss': 1.2437, 'grad_norm': 6.012956619262695, 'learning_rate': 1.9302320959231997e-05, 'epoch': 0.49}

 16%|█▋        | 420/2562 [21:20<1:43:34,  2.90s/it]
 16%|█▋        | 421/2562 [21:23<1:42:43,  2.88s/it]
 16%|█▋        | 422/2562 [21:26<1:42:49,  2.88s/it]
 17%|█▋        | 423/2562 [21:29<1:43:54,  2.91s/it]
 17%|█▋        | 424/2562 [21:32<1:43:20,  2.90s/it]
 17%|█▋        | 425/2562 [21:35<1:42:20,  2.87s/it]
 17%|█▋        | 426/2562 [21:38<1:43:44,  2.91s/it]
 17%|█▋        | 427/2562 [21:41<1:46:15,  2.99s/it]
 17%|█▋        | 428/2562 [21:44<1:46:13,  2.99s/it]
 17%|█▋        | 429/2562 [21:47<1:44:30,  2.94s/it]
 17%|█▋        | 430/2562 [21:50<1:45:06,  2.96s/it]
                                                    
{'loss': 1.2437, 'grad_norm': 6.023150444030762, 'learning_rate': 1.9254161851955587e-05, 'epoch': 0.5}

 17%|█▋        | 430/2562 [21:50<1:45:06,  2.96s/it]
 17%|█▋        | 431/2562 [21:53<1:46:43,  3.00s/it]
 17%|█▋        | 432/2562 [21:56<1:44:52,  2.95s/it]
 17%|█▋        | 433/2562 [21:59<1:45:06,  2.96s/it]
 17%|█▋        | 434/2562 [22:02<1:44:06,  2.94s/it]
 17%|█▋        | 435/2562 [22:04<1:43:03,  2.91s/it]
 17%|█▋        | 436/2562 [22:07<1:43:29,  2.92s/it]
 17%|█▋        | 437/2562 [22:10<1:43:32,  2.92s/it]
 17%|█▋        | 438/2562 [22:13<1:44:22,  2.95s/it]
 17%|█▋        | 439/2562 [22:16<1:42:54,  2.91s/it]
 17%|█▋        | 440/2562 [22:19<1:45:20,  2.98s/it]
                                                    
{'loss': 1.2227, 'grad_norm': 5.310694694519043, 'learning_rate': 1.920445981339708e-05, 'epoch': 0.51}

 17%|█▋        | 440/2562 [22:19<1:45:20,  2.98s/it]
 17%|█▋        | 441/2562 [22:22<1:43:05,  2.92s/it]
 17%|█▋        | 442/2562 [22:25<1:43:35,  2.93s/it]
 17%|█▋        | 443/2562 [22:28<1:44:40,  2.96s/it]
 17%|█▋        | 444/2562 [22:31<1:42:44,  2.91s/it]
 17%|█▋        | 445/2562 [22:34<1:42:15,  2.90s/it]
 17%|█▋        | 446/2562 [22:37<1:42:37,  2.91s/it]
 17%|█▋        | 447/2562 [22:39<1:42:28,  2.91s/it]
 17%|█▋        | 448/2562 [22:42<1:41:45,  2.89s/it]
 18%|█▊        | 449/2562 [22:45<1:42:34,  2.91s/it]
 18%|█▊        | 450/2562 [22:48<1:42:08,  2.90s/it]
                                                    
{'loss': 1.2053, 'grad_norm': 5.454719543457031, 'learning_rate': 1.9153223130296125e-05, 'epoch': 0.53}

 18%|█▊        | 450/2562 [22:48<1:42:08,  2.90s/it]
 18%|█▊        | 451/2562 [22:51<1:40:45,  2.86s/it]
 18%|█▊        | 452/2562 [22:54<1:41:14,  2.88s/it]
 18%|█▊        | 453/2562 [22:57<1:40:01,  2.85s/it]
 18%|█▊        | 454/2562 [22:59<1:39:49,  2.84s/it]
 18%|█▊        | 455/2562 [23:03<1:42:09,  2.91s/it]
 18%|█▊        | 456/2562 [23:05<1:40:29,  2.86s/it]
 18%|█▊        | 457/2562 [23:08<1:42:01,  2.91s/it]
 18%|█▊        | 458/2562 [23:11<1:42:27,  2.92s/it]
 18%|█▊        | 459/2562 [23:14<1:40:29,  2.87s/it]
 18%|█▊        | 460/2562 [23:17<1:40:53,  2.88s/it]
                                                    
{'loss': 1.2135, 'grad_norm': 5.662570476531982, 'learning_rate': 1.9100460345261175e-05, 'epoch': 0.54}

 18%|█▊        | 460/2562 [23:17<1:40:53,  2.88s/it]
 18%|█▊        | 461/2562 [23:20<1:41:06,  2.89s/it]
 18%|█▊        | 462/2562 [23:23<1:40:13,  2.86s/it]
 18%|█▊        | 463/2562 [23:26<1:40:42,  2.88s/it]
 18%|█▊        | 464/2562 [23:28<1:41:16,  2.90s/it]
 18%|█▊        | 465/2562 [23:31<1:39:42,  2.85s/it]
 18%|█▊        | 466/2562 [23:34<1:40:09,  2.87s/it]
 18%|█▊        | 467/2562 [23:37<1:39:56,  2.86s/it]
 18%|█▊        | 468/2562 [23:40<1:40:01,  2.87s/it]
 18%|█▊        | 469/2562 [23:43<1:41:23,  2.91s/it]
 18%|█▊        | 470/2562 [23:46<1:39:50,  2.86s/it]
                                                    
{'loss': 1.2195, 'grad_norm': 5.714497089385986, 'learning_rate': 1.9046180255345142e-05, 'epoch': 0.55}

 18%|█▊        | 470/2562 [23:46<1:39:50,  2.86s/it]
 18%|█▊        | 471/2562 [23:48<1:38:36,  2.83s/it]
 18%|█▊        | 472/2562 [23:51<1:39:51,  2.87s/it]
 18%|█▊        | 473/2562 [23:54<1:38:54,  2.84s/it]
 19%|█▊        | 474/2562 [23:57<1:39:26,  2.86s/it]
 19%|█▊        | 475/2562 [24:00<1:40:04,  2.88s/it]
 19%|█▊        | 476/2562 [24:03<1:40:56,  2.90s/it]
 19%|█▊        | 477/2562 [24:06<1:40:43,  2.90s/it]
 19%|█▊        | 478/2562 [24:09<1:41:33,  2.92s/it]
 19%|█▊        | 479/2562 [24:12<1:40:20,  2.89s/it]
 19%|█▊        | 480/2562 [24:15<1:40:56,  2.91s/it]
                                                    
{'loss': 1.1807, 'grad_norm': 5.568943500518799, 'learning_rate': 1.899039191057872e-05, 'epoch': 0.56}

 19%|█▊        | 480/2562 [24:15<1:40:56,  2.91s/it]
 19%|█▉        | 481/2562 [24:17<1:41:24,  2.92s/it]
 19%|█▉        | 482/2562 [24:20<1:39:35,  2.87s/it]
 19%|█▉        | 483/2562 [24:23<1:41:23,  2.93s/it]
 19%|█▉        | 484/2562 [24:26<1:40:18,  2.90s/it]
 19%|█▉        | 485/2562 [24:29<1:38:37,  2.85s/it]
 19%|█▉        | 486/2562 [24:32<1:40:48,  2.91s/it]
 19%|█▉        | 487/2562 [24:35<1:41:42,  2.94s/it]
 19%|█▉        | 488/2562 [24:38<1:40:49,  2.92s/it]
 19%|█▉        | 489/2562 [24:41<1:41:40,  2.94s/it]
 19%|█▉        | 490/2562 [24:44<1:42:03,  2.96s/it]
                                                    
{'loss': 1.1759, 'grad_norm': 4.761386871337891, 'learning_rate': 1.8933104612461454e-05, 'epoch': 0.57}

 19%|█▉        | 490/2562 [24:44<1:42:03,  2.96s/it]
 19%|█▉        | 491/2562 [24:47<1:40:41,  2.92s/it]
 19%|█▉        | 492/2562 [24:50<1:42:34,  2.97s/it]
 19%|█▉        | 493/2562 [24:53<1:41:04,  2.93s/it]
 19%|█▉        | 494/2562 [24:55<1:39:22,  2.88s/it]
 19%|█▉        | 495/2562 [24:58<1:39:37,  2.89s/it]
 19%|█▉        | 496/2562 [25:01<1:38:49,  2.87s/it]
 19%|█▉        | 497/2562 [25:04<1:37:17,  2.83s/it]
 19%|█▉        | 498/2562 [25:07<1:40:24,  2.92s/it]
 19%|█▉        | 499/2562 [25:10<1:40:49,  2.93s/it]
 20%|█▉        | 500/2562 [25:13<1:38:45,  2.87s/it]
                                                    
{'loss': 1.1582, 'grad_norm': 5.17278528213501, 'learning_rate': 1.8874327912410945e-05, 'epoch': 0.58}

 20%|█▉        | 500/2562 [25:13<1:38:45,  2.87s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.64it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.71it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.22it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.89it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.55it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.74it/s]

 11%|█         | 15/140 [00:01<00:12,  9.77it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.75it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.78it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.57it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.45it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.60it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.39it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.35it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.49it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.34it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.05it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.12it/s]

 20%|██        | 28/140 [00:02<00:11,  9.50it/s]

 21%|██        | 29/140 [00:02<00:11,  9.35it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.05it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.06it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.26it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.72it/s]

 25%|██▌       | 35/140 [00:03<00:10,  9.77it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.75it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.79it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.91it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.90it/s]

 29%|██▉       | 41/140 [00:04<00:10,  9.88it/s]

 30%|███       | 42/140 [00:04<00:09,  9.86it/s]

 31%|███       | 43/140 [00:04<00:09,  9.88it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.86it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.85it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.57it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.21it/s]

 34%|███▍      | 48/140 [00:04<00:10,  9.19it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.11it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.21it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.40it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.40it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.43it/s]

 40%|████      | 56/140 [00:05<00:08,  9.70it/s]

 41%|████      | 57/140 [00:05<00:08,  9.49it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.40it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.61it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.80it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.60it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.54it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.54it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.51it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.72it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.77it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.84it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.86it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.83it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.82it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.59it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.19it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.29it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.28it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.18it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.06it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.07it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.30it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.20it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.16it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.05it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.08it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.98it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.07it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.24it/s]

 66%|██████▋   | 93/140 [00:09<00:05,  9.40it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.23it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.01it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.09it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.31it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.18it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.24it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.40it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.25it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.34it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.53it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.76it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.63it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.52it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.61it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.65it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.66it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.83it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.85it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.54it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.43it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.55it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.40it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.10it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.07it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.23it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.21it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.36it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.26it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.02it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.80it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.56it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.67it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  8.99it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.48it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.60it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.61it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.43it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.46it/s]

100%|██████████| 140/140 [00:14<00:00,  8.04it/s]
                                                    

                                                 
{'eval_loss': 1.210729956626892, 'eval_runtime': 15.1584, 'eval_samples_per_second': 18.472, 'eval_steps_per_second': 9.236, 'epoch': 0.58}

 20%|█▉        | 500/2562 [25:28<1:38:45,  2.87s/it]

100%|██████████| 140/140 [00:14<00:00,  8.04it/s]

                                                 /opt/conda/lib/python3.11/site-packages/peft/utils/save_and_load.py:372: UserWarning: Could not find a config file in  - will assume that the vocabulary was not modified.
  warnings.warn(

 20%|█▉        | 501/2562 [25:32<4:33:06,  7.95s/it]
 20%|█▉        | 502/2562 [25:35<3:39:26,  6.39s/it]
 20%|█▉        | 503/2562 [25:38<3:01:57,  5.30s/it]
 20%|█▉        | 504/2562 [25:41<2:38:28,  4.62s/it]
 20%|█▉        | 505/2562 [25:44<2:21:15,  4.12s/it]
 20%|█▉        | 506/2562 [25:47<2:09:02,  3.77s/it]
 20%|█▉        | 507/2562 [25:50<1:59:39,  3.49s/it]
 20%|█▉        | 508/2562 [25:53<1:52:50,  3.30s/it]
 20%|█▉        | 509/2562 [25:56<1:49:37,  3.20s/it]
 20%|█▉        | 510/2562 [25:58<1:45:42,  3.09s/it]
                                                    
{'loss': 1.1977, 'grad_norm': 5.511214256286621, 'learning_rate': 1.881407161017033e-05, 'epoch': 0.6}

 20%|█▉        | 510/2562 [25:58<1:45:42,  3.09s/it]
 20%|█▉        | 511/2562 [26:01<1:41:45,  2.98s/it]
 20%|█▉        | 512/2562 [26:04<1:42:54,  3.01s/it]
 20%|██        | 513/2562 [26:07<1:42:28,  3.00s/it]
 20%|██        | 514/2562 [26:10<1:40:10,  2.93s/it]
 20%|██        | 515/2562 [26:13<1:42:23,  3.00s/it]
 20%|██        | 516/2562 [26:16<1:40:31,  2.95s/it]
 20%|██        | 517/2562 [26:19<1:40:06,  2.94s/it]
 20%|██        | 518/2562 [26:22<1:39:33,  2.92s/it]
 20%|██        | 519/2562 [26:24<1:37:31,  2.86s/it]
 20%|██        | 520/2562 [26:27<1:37:31,  2.87s/it]
                                                    
{'loss': 1.1529, 'grad_norm': 5.508382320404053, 'learning_rate': 1.875234575217441e-05, 'epoch': 0.61}

 20%|██        | 520/2562 [26:27<1:37:31,  2.87s/it]
 20%|██        | 521/2562 [26:30<1:39:57,  2.94s/it]
 20%|██        | 522/2562 [26:33<1:39:15,  2.92s/it]
 20%|██        | 523/2562 [26:36<1:40:26,  2.96s/it]
 20%|██        | 524/2562 [26:39<1:38:43,  2.91s/it]
 20%|██        | 525/2562 [26:42<1:38:30,  2.90s/it]
 21%|██        | 526/2562 [26:45<1:39:34,  2.93s/it]
 21%|██        | 527/2562 [26:48<1:38:39,  2.91s/it]
 21%|██        | 528/2562 [26:51<1:38:23,  2.90s/it]
 21%|██        | 529/2562 [26:54<1:38:30,  2.91s/it]
 21%|██        | 530/2562 [26:57<1:38:09,  2.90s/it]
                                                    
{'loss': 1.1478, 'grad_norm': 4.957241535186768, 'learning_rate': 1.8689160629874622e-05, 'epoch': 0.62}

 21%|██        | 530/2562 [26:57<1:38:09,  2.90s/it]
 21%|██        | 531/2562 [27:00<1:39:27,  2.94s/it]
 21%|██        | 532/2562 [27:02<1:37:08,  2.87s/it]
 21%|██        | 533/2562 [27:05<1:40:04,  2.96s/it]
 21%|██        | 534/2562 [27:08<1:38:38,  2.92s/it]
 21%|██        | 535/2562 [27:11<1:41:16,  3.00s/it]
 21%|██        | 536/2562 [27:14<1:39:00,  2.93s/it]
 21%|██        | 537/2562 [27:17<1:38:39,  2.92s/it]
 21%|██        | 538/2562 [27:20<1:38:40,  2.93s/it]
 21%|██        | 539/2562 [27:23<1:37:32,  2.89s/it]
 21%|██        | 540/2562 [27:26<1:36:48,  2.87s/it]
                                                    
{'loss': 1.1242, 'grad_norm': 5.256776809692383, 'learning_rate': 1.8624526778023142e-05, 'epoch': 0.63}

 21%|██        | 540/2562 [27:26<1:36:48,  2.87s/it]
 21%|██        | 541/2562 [27:29<1:37:46,  2.90s/it]
 21%|██        | 542/2562 [27:32<1:37:42,  2.90s/it]
 21%|██        | 543/2562 [27:34<1:36:31,  2.87s/it]
 21%|██        | 544/2562 [27:37<1:38:31,  2.93s/it]
 21%|██▏       | 545/2562 [27:40<1:38:34,  2.93s/it]
 21%|██▏       | 546/2562 [27:43<1:37:50,  2.91s/it]
 21%|██▏       | 547/2562 [27:46<1:38:07,  2.92s/it]
 21%|██▏       | 548/2562 [27:49<1:37:54,  2.92s/it]
 21%|██▏       | 549/2562 [27:52<1:39:11,  2.96s/it]
 21%|██▏       | 550/2562 [27:55<1:39:29,  2.97s/it]
                                                    
{'loss': 1.1711, 'grad_norm': 5.197915554046631, 'learning_rate': 1.855845497291646e-05, 'epoch': 0.64}

 21%|██▏       | 550/2562 [27:55<1:39:29,  2.97s/it]
 22%|██▏       | 551/2562 [27:58<1:38:47,  2.95s/it]
 22%|██▏       | 552/2562 [28:01<1:36:58,  2.89s/it]
 22%|██▏       | 553/2562 [28:04<1:38:20,  2.94s/it]
 22%|██▏       | 554/2562 [28:07<1:39:38,  2.98s/it]
 22%|██▏       | 555/2562 [28:10<1:40:10,  3.00s/it]
 22%|██▏       | 556/2562 [28:13<1:39:18,  2.97s/it]
 22%|██▏       | 557/2562 [28:16<1:37:46,  2.93s/it]
 22%|██▏       | 558/2562 [28:19<1:39:53,  2.99s/it]
 22%|██▏       | 559/2562 [28:22<1:40:29,  3.01s/it]
 22%|██▏       | 560/2562 [28:25<1:38:44,  2.96s/it]
                                                    
{'loss': 1.1033, 'grad_norm': 5.34417724609375, 'learning_rate': 1.8490956230598668e-05, 'epoch': 0.66}

 22%|██▏       | 560/2562 [28:25<1:38:44,  2.96s/it]
 22%|██▏       | 561/2562 [28:28<1:38:47,  2.96s/it]
 22%|██▏       | 562/2562 [28:31<1:38:36,  2.96s/it]
 22%|██▏       | 563/2562 [28:34<1:37:58,  2.94s/it]
 22%|██▏       | 564/2562 [28:36<1:38:06,  2.95s/it]
 22%|██▏       | 565/2562 [28:40<1:38:47,  2.97s/it]
 22%|██▏       | 566/2562 [28:42<1:37:23,  2.93s/it]
 22%|██▏       | 567/2562 [28:45<1:36:37,  2.91s/it]
 22%|██▏       | 568/2562 [28:48<1:38:30,  2.96s/it]
 22%|██▏       | 569/2562 [28:51<1:37:10,  2.93s/it]
 22%|██▏       | 570/2562 [28:54<1:36:46,  2.91s/it]
                                                    
{'loss': 1.141, 'grad_norm': 5.578720569610596, 'learning_rate': 1.842204180502476e-05, 'epoch': 0.67}

 22%|██▏       | 570/2562 [28:54<1:36:46,  2.91s/it]
 22%|██▏       | 571/2562 [28:57<1:38:33,  2.97s/it]
 22%|██▏       | 572/2562 [29:00<1:37:59,  2.95s/it]
 22%|██▏       | 573/2562 [29:03<1:36:37,  2.91s/it]
 22%|██▏       | 574/2562 [29:06<1:37:58,  2.96s/it]
 22%|██▏       | 575/2562 [29:09<1:37:19,  2.94s/it]
 22%|██▏       | 576/2562 [29:12<1:36:56,  2.93s/it]
 23%|██▎       | 577/2562 [29:15<1:37:29,  2.95s/it]
 23%|██▎       | 578/2562 [29:18<1:35:53,  2.90s/it]
 23%|██▎       | 579/2562 [29:20<1:35:47,  2.90s/it]
 23%|██▎       | 580/2562 [29:23<1:37:12,  2.94s/it]
                                                    
{'loss': 1.1261, 'grad_norm': 5.381358623504639, 'learning_rate': 1.8351723186184295e-05, 'epoch': 0.68}

 23%|██▎       | 580/2562 [29:23<1:37:12,  2.94s/it]
 23%|██▎       | 581/2562 [29:26<1:36:54,  2.94s/it]
 23%|██▎       | 582/2562 [29:29<1:37:10,  2.94s/it]
 23%|██▎       | 583/2562 [29:32<1:36:26,  2.92s/it]
 23%|██▎       | 584/2562 [29:35<1:35:50,  2.91s/it]
 23%|██▎       | 585/2562 [29:38<1:35:59,  2.91s/it]
 23%|██▎       | 586/2562 [29:41<1:36:29,  2.93s/it]
 23%|██▎       | 587/2562 [29:44<1:35:57,  2.92s/it]
 23%|██▎       | 588/2562 [29:47<1:35:08,  2.89s/it]
 23%|██▎       | 589/2562 [29:50<1:36:01,  2.92s/it]
 23%|██▎       | 590/2562 [29:53<1:37:00,  2.95s/it]
                                                    
{'loss': 1.1174, 'grad_norm': 5.501519203186035, 'learning_rate': 1.828001209818567e-05, 'epoch': 0.69}

 23%|██▎       | 590/2562 [29:53<1:37:00,  2.95s/it]
 23%|██▎       | 591/2562 [29:56<1:38:36,  3.00s/it]
 23%|██▎       | 592/2562 [29:59<1:37:03,  2.96s/it]
 23%|██▎       | 593/2562 [30:02<1:35:50,  2.92s/it]
 23%|██▎       | 594/2562 [30:05<1:37:05,  2.96s/it]
 23%|██▎       | 595/2562 [30:07<1:34:51,  2.89s/it]
 23%|██▎       | 596/2562 [30:10<1:34:09,  2.87s/it]
 23%|██▎       | 597/2562 [30:13<1:35:57,  2.93s/it]
 23%|██▎       | 598/2562 [30:16<1:36:16,  2.94s/it]
 23%|██▎       | 599/2562 [30:19<1:34:54,  2.90s/it]
 23%|██▎       | 600/2562 [30:22<1:36:24,  2.95s/it]
                                                    
{'loss': 1.1387, 'grad_norm': 6.146886348724365, 'learning_rate': 1.82069204973014e-05, 'epoch': 0.7}

 23%|██▎       | 600/2562 [30:22<1:36:24,  2.95s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.64it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.78it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.24it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.90it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.54it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.74it/s]

 10%|█         | 14/140 [00:01<00:12,  9.78it/s]

 11%|█         | 15/140 [00:01<00:12,  9.83it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.83it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.70it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.64it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.57it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.47it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.56it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.41it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.14it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.19it/s]

 20%|██        | 28/140 [00:02<00:11,  9.51it/s]

 21%|██        | 29/140 [00:02<00:11,  9.35it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.05it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.06it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.30it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.76it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.83it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.92it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.95it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.94it/s]

 30%|███       | 42/140 [00:04<00:09,  9.93it/s]

 31%|███       | 43/140 [00:04<00:09,  9.94it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.91it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.87it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.63it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.28it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.26it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.20it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.27it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.46it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.47it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.46it/s]

 40%|████      | 56/140 [00:05<00:08,  9.70it/s]

 41%|████      | 57/140 [00:05<00:08,  9.48it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.38it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.48it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.55it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.80it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.60it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.54it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.53it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.50it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.74it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.81it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.85it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.88it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.87it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.87it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.61it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.22it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.34it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.30it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.22it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.10it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.10it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.30it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.21it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.19it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.10it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.11it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.00it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.05it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.25it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.41it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.25it/s]

 68%|██████▊   | 95/140 [00:09<00:05,  8.99it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.12it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.29it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.34it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.50it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.32it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.38it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.70it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.72it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.57it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.46it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.56it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.59it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.63it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.80it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.81it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.50it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.39it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.52it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.27it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.40it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.10it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.07it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.21it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.39it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.30it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.08it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.85it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.60it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.72it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.05it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.31it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.68it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.70it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.50it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.51it/s]

100%|██████████| 140/140 [00:14<00:00,  8.12it/s]
                                                    

                                                 
{'eval_loss': 1.1566864252090454, 'eval_runtime': 15.1228, 'eval_samples_per_second': 18.515, 'eval_steps_per_second': 9.258, 'epoch': 0.7}

 23%|██▎       | 600/2562 [30:37<1:36:24,  2.95s/it]

100%|██████████| 140/140 [00:14<00:00,  8.12it/s]

                                                 
 23%|██▎       | 601/2562 [30:40<4:02:28,  7.42s/it]
 23%|██▎       | 602/2562 [30:43<3:19:51,  6.12s/it]
 24%|██▎       | 603/2562 [30:46<2:46:53,  5.11s/it]
 24%|██▎       | 604/2562 [30:49<2:24:39,  4.43s/it]
 24%|██▎       | 605/2562 [30:52<2:10:07,  3.99s/it]
 24%|██▎       | 606/2562 [30:55<2:00:53,  3.71s/it]
 24%|██▎       | 607/2562 [30:57<1:52:24,  3.45s/it]
 24%|██▎       | 608/2562 [31:01<1:49:51,  3.37s/it]
 24%|██▍       | 609/2562 [31:04<1:45:37,  3.25s/it]
 24%|██▍       | 610/2562 [31:06<1:41:35,  3.12s/it]
                                                    
{'loss': 1.1224, 'grad_norm': 5.354383945465088, 'learning_rate': 1.813246056997465e-05, 'epoch': 0.71}

 24%|██▍       | 610/2562 [31:06<1:41:35,  3.12s/it]
 24%|██▍       | 611/2562 [31:09<1:40:47,  3.10s/it]
 24%|██▍       | 612/2562 [31:12<1:37:24,  3.00s/it]
 24%|██▍       | 613/2562 [31:15<1:37:35,  3.00s/it]
 24%|██▍       | 614/2562 [31:18<1:36:43,  2.98s/it]
 24%|██▍       | 615/2562 [31:21<1:36:12,  2.96s/it]
 24%|██▍       | 616/2562 [31:24<1:35:56,  2.96s/it]
 24%|██▍       | 617/2562 [31:27<1:36:51,  2.99s/it]
 24%|██▍       | 618/2562 [31:30<1:35:42,  2.95s/it]
 24%|██▍       | 619/2562 [31:33<1:36:56,  2.99s/it]
 24%|██▍       | 620/2562 [31:36<1:35:05,  2.94s/it]
                                                    
{'loss': 1.1018, 'grad_norm': 4.961958408355713, 'learning_rate': 1.8056644730787412e-05, 'epoch': 0.73}

 24%|██▍       | 620/2562 [31:36<1:35:05,  2.94s/it]
 24%|██▍       | 621/2562 [31:39<1:35:03,  2.94s/it]
 24%|██▍       | 622/2562 [31:42<1:34:56,  2.94s/it]
 24%|██▍       | 623/2562 [31:44<1:32:45,  2.87s/it]
 24%|██▍       | 624/2562 [31:47<1:33:25,  2.89s/it]
 24%|██▍       | 625/2562 [31:50<1:34:46,  2.94s/it]
 24%|██▍       | 626/2562 [31:53<1:34:45,  2.94s/it]
 24%|██▍       | 627/2562 [31:56<1:33:34,  2.90s/it]
 25%|██▍       | 628/2562 [31:59<1:34:46,  2.94s/it]
 25%|██▍       | 629/2562 [32:02<1:35:54,  2.98s/it]
 25%|██▍       | 630/2562 [32:05<1:36:23,  2.99s/it]
                                                    
{'loss': 1.1163, 'grad_norm': 5.070528984069824, 'learning_rate': 1.797948562039066e-05, 'epoch': 0.74}

 25%|██▍       | 630/2562 [32:05<1:36:23,  2.99s/it]
 25%|██▍       | 631/2562 [32:08<1:35:27,  2.97s/it]
 25%|██▍       | 632/2562 [32:11<1:33:38,  2.91s/it]
 25%|██▍       | 633/2562 [32:14<1:32:58,  2.89s/it]
 25%|██▍       | 634/2562 [32:17<1:32:59,  2.89s/it]
 25%|██▍       | 635/2562 [32:20<1:33:05,  2.90s/it]
 25%|██▍       | 636/2562 [32:23<1:35:19,  2.97s/it]
 25%|██▍       | 637/2562 [32:26<1:35:11,  2.97s/it]
 25%|██▍       | 638/2562 [32:29<1:36:08,  3.00s/it]
 25%|██▍       | 639/2562 [32:32<1:35:54,  2.99s/it]
 25%|██▍       | 640/2562 [32:35<1:34:28,  2.95s/it]
                                                    
{'loss': 1.1013, 'grad_norm': 5.25916862487793, 'learning_rate': 1.7900996103396772e-05, 'epoch': 0.75}

 25%|██▍       | 640/2562 [32:35<1:34:28,  2.95s/it]
 25%|██▌       | 641/2562 [32:37<1:33:25,  2.92s/it]
 25%|██▌       | 642/2562 [32:40<1:33:31,  2.92s/it]
 25%|██▌       | 643/2562 [32:43<1:33:58,  2.94s/it]
 25%|██▌       | 644/2562 [32:46<1:34:31,  2.96s/it]
 25%|██▌       | 645/2562 [32:49<1:33:44,  2.93s/it]
 25%|██▌       | 646/2562 [32:52<1:31:55,  2.88s/it]
 25%|██▌       | 647/2562 [32:55<1:31:29,  2.87s/it]
 25%|██▌       | 648/2562 [32:58<1:32:14,  2.89s/it]
 25%|██▌       | 649/2562 [33:01<1:32:18,  2.90s/it]
 25%|██▌       | 650/2562 [33:04<1:32:12,  2.89s/it]
                                                    
{'loss': 1.102, 'grad_norm': 6.132543563842773, 'learning_rate': 1.7821189266234647e-05, 'epoch': 0.76}

 25%|██▌       | 650/2562 [33:04<1:32:12,  2.89s/it]
 25%|██▌       | 651/2562 [33:07<1:32:27,  2.90s/it]
 25%|██▌       | 652/2562 [33:09<1:32:34,  2.91s/it]
 25%|██▌       | 653/2562 [33:12<1:32:40,  2.91s/it]
 26%|██▌       | 654/2562 [33:15<1:33:40,  2.95s/it]
 26%|██▌       | 655/2562 [33:18<1:32:59,  2.93s/it]
 26%|██▌       | 656/2562 [33:21<1:32:04,  2.90s/it]
 26%|██▌       | 657/2562 [33:24<1:33:38,  2.95s/it]
 26%|██▌       | 658/2562 [33:27<1:33:43,  2.95s/it]
 26%|██▌       | 659/2562 [33:30<1:32:02,  2.90s/it]
 26%|██▌       | 660/2562 [33:33<1:32:53,  2.93s/it]
                                                    
{'loss': 1.0611, 'grad_norm': 5.370880126953125, 'learning_rate': 1.7740078414967817e-05, 'epoch': 0.77}

 26%|██▌       | 660/2562 [33:33<1:32:53,  2.93s/it]
 26%|██▌       | 661/2562 [33:36<1:32:44,  2.93s/it]
 26%|██▌       | 662/2562 [33:39<1:34:03,  2.97s/it]
 26%|██▌       | 663/2562 [33:42<1:33:34,  2.96s/it]
 26%|██▌       | 664/2562 [33:45<1:32:23,  2.92s/it]
 26%|██▌       | 665/2562 [33:48<1:31:57,  2.91s/it]
 26%|██▌       | 666/2562 [33:50<1:31:32,  2.90s/it]
 26%|██▌       | 667/2562 [33:53<1:31:29,  2.90s/it]
 26%|██▌       | 668/2562 [33:56<1:31:57,  2.91s/it]
 26%|██▌       | 669/2562 [33:59<1:32:20,  2.93s/it]
 26%|██▌       | 670/2562 [34:02<1:32:05,  2.92s/it]
                                                    
{'loss': 1.0789, 'grad_norm': 5.0857086181640625, 'learning_rate': 1.7657677073075968e-05, 'epoch': 0.78}

 26%|██▌       | 670/2562 [34:02<1:32:05,  2.92s/it]
 26%|██▌       | 671/2562 [34:05<1:31:00,  2.89s/it]
 26%|██▌       | 672/2562 [34:08<1:32:05,  2.92s/it]
 26%|██▋       | 673/2562 [34:11<1:34:44,  3.01s/it]
 26%|██▋       | 674/2562 [34:14<1:32:33,  2.94s/it]
 26%|██▋       | 675/2562 [34:17<1:33:00,  2.96s/it]
 26%|██▋       | 676/2562 [34:20<1:31:44,  2.92s/it]
 26%|██▋       | 677/2562 [34:23<1:30:48,  2.89s/it]
 26%|██▋       | 678/2562 [34:26<1:32:46,  2.95s/it]
 27%|██▋       | 679/2562 [34:28<1:30:55,  2.90s/it]
 27%|██▋       | 680/2562 [34:31<1:31:09,  2.91s/it]
                                                    
{'loss': 1.1036, 'grad_norm': 5.835397243499756, 'learning_rate': 1.7573998979200163e-05, 'epoch': 0.8}

 27%|██▋       | 680/2562 [34:31<1:31:09,  2.91s/it]
 27%|██▋       | 681/2562 [34:34<1:31:07,  2.91s/it]
 27%|██▋       | 682/2562 [34:37<1:30:42,  2.90s/it]
 27%|██▋       | 683/2562 [34:40<1:30:52,  2.90s/it]
 27%|██▋       | 684/2562 [34:43<1:30:40,  2.90s/it]
 27%|██▋       | 685/2562 [34:46<1:30:56,  2.91s/it]
 27%|██▋       | 686/2562 [34:49<1:31:36,  2.93s/it]
 27%|██▋       | 687/2562 [34:52<1:30:47,  2.91s/it]
 27%|██▋       | 688/2562 [34:54<1:29:27,  2.86s/it]
 27%|██▋       | 689/2562 [34:57<1:29:17,  2.86s/it]
 27%|██▋       | 690/2562 [35:00<1:29:42,  2.88s/it]
                                                    
{'loss': 1.1158, 'grad_norm': 6.6026930809021, 'learning_rate': 1.7489058084852247e-05, 'epoch': 0.81}

 27%|██▋       | 690/2562 [35:00<1:29:42,  2.88s/it]
 27%|██▋       | 691/2562 [35:03<1:29:27,  2.87s/it]
 27%|██▋       | 692/2562 [35:06<1:30:14,  2.90s/it]
 27%|██▋       | 693/2562 [35:09<1:29:01,  2.86s/it]
 27%|██▋       | 694/2562 [35:12<1:30:06,  2.89s/it]
 27%|██▋       | 695/2562 [35:15<1:30:00,  2.89s/it]
 27%|██▋       | 696/2562 [35:17<1:28:44,  2.85s/it]
 27%|██▋       | 697/2562 [35:20<1:28:17,  2.84s/it]
 27%|██▋       | 698/2562 [35:23<1:29:28,  2.88s/it]
 27%|██▋       | 699/2562 [35:26<1:30:26,  2.91s/it]
 27%|██▋       | 700/2562 [35:29<1:29:28,  2.88s/it]
                                                    
{'loss': 1.0645, 'grad_norm': 5.48411750793457, 'learning_rate': 1.7402868552088724e-05, 'epoch': 0.82}

 27%|██▋       | 700/2562 [35:29<1:29:28,  2.88s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.59it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.69it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.19it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.79it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.49it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.68it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.72it/s]

 10%|█         | 14/140 [00:01<00:12,  9.75it/s]

 11%|█         | 15/140 [00:01<00:12,  9.79it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.79it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.83it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.61it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.54it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.65it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.40it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.36it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.48it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.35it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.01it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.09it/s]

 20%|██        | 28/140 [00:02<00:11,  9.44it/s]

 21%|██        | 29/140 [00:02<00:11,  9.30it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.00it/s]

 22%|██▏       | 31/140 [00:03<00:12,  8.99it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.21it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.65it/s]

 25%|██▌       | 35/140 [00:03<00:10,  9.73it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.76it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.79it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.85it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.92it/s]

 29%|██▉       | 41/140 [00:04<00:10,  9.90it/s]

 31%|███       | 43/140 [00:04<00:09,  9.95it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.91it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.88it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.62it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.28it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.24it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.15it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.21it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.39it/s]

 37%|███▋      | 52/140 [00:05<00:09,  9.56it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.37it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.35it/s]

 40%|████      | 56/140 [00:05<00:08,  9.67it/s]

 41%|████      | 57/140 [00:05<00:08,  9.46it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.35it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.45it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.56it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.78it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.59it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.53it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.50it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.48it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.71it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.76it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.77it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.80it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.83it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.84it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.81it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.80it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.56it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.18it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.32it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.26it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.18it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.09it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.11it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.31it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.19it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.16it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.06it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.08it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.97it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.04it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.24it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.42it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.26it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.01it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.10it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.27it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.32it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.45it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.28it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.34it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.51it/s]

 75%|███████▌  | 105/140 [00:11<00:03,  9.75it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.62it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.49it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.58it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.62it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.64it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.83it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.83it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.52it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.41it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.54it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.27it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.40it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.10it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.07it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.22it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.39it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.29it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.07it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.83it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.57it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.70it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  9.02it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.26it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.63it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.67it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.51it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.51it/s]

100%|██████████| 140/140 [00:14<00:00,  8.11it/s]
                                                    

                                                 
{'eval_loss': 1.1051727533340454, 'eval_runtime': 15.165, 'eval_samples_per_second': 18.464, 'eval_steps_per_second': 9.232, 'epoch': 0.82}

 27%|██▋       | 700/2562 [35:44<1:29:28,  2.88s/it]

100%|██████████| 140/140 [00:14<00:00,  8.11it/s]

                                                 
 27%|██▋       | 701/2562 [35:47<3:53:46,  7.54s/it]
 27%|██▋       | 702/2562 [35:50<3:10:03,  6.13s/it]
 27%|██▋       | 703/2562 [35:53<2:40:53,  5.19s/it]
 27%|██▋       | 704/2562 [35:56<2:20:46,  4.55s/it]
 28%|██▊       | 705/2562 [35:59<2:05:21,  4.05s/it]
 28%|██▊       | 706/2562 [36:02<1:52:45,  3.65s/it]
 28%|██▊       | 707/2562 [36:05<1:45:57,  3.43s/it]
 28%|██▊       | 708/2562 [36:08<1:41:00,  3.27s/it]
 28%|██▊       | 709/2562 [36:10<1:36:14,  3.12s/it]
 28%|██▊       | 710/2562 [36:13<1:35:00,  3.08s/it]
                                                    
{'loss': 1.0917, 'grad_norm': 5.765743732452393, 'learning_rate': 1.7315444751149533e-05, 'epoch': 0.83}

 28%|██▊       | 710/2562 [36:13<1:35:00,  3.08s/it]
 28%|██▊       | 711/2562 [36:16<1:32:29,  3.00s/it]
 28%|██▊       | 712/2562 [36:19<1:31:09,  2.96s/it]
 28%|██▊       | 713/2562 [36:22<1:32:41,  3.01s/it]
 28%|██▊       | 714/2562 [36:25<1:30:39,  2.94s/it]
 28%|██▊       | 715/2562 [36:28<1:29:38,  2.91s/it]
 28%|██▊       | 716/2562 [36:31<1:29:45,  2.92s/it]
 28%|██▊       | 717/2562 [36:34<1:29:01,  2.89s/it]
 28%|██▊       | 718/2562 [36:37<1:29:02,  2.90s/it]
 28%|██▊       | 719/2562 [36:40<1:29:47,  2.92s/it]
 28%|██▊       | 720/2562 [36:42<1:29:29,  2.92s/it]
                                                    
{'loss': 1.08, 'grad_norm': 5.12139368057251, 'learning_rate': 1.722680125806214e-05, 'epoch': 0.84}

 28%|██▊       | 720/2562 [36:42<1:29:29,  2.92s/it]
 28%|██▊       | 721/2562 [36:45<1:29:28,  2.92s/it]
 28%|██▊       | 722/2562 [36:48<1:28:48,  2.90s/it]
 28%|██▊       | 723/2562 [36:51<1:27:35,  2.86s/it]
 28%|██▊       | 724/2562 [36:54<1:29:56,  2.94s/it]
 28%|██▊       | 725/2562 [36:57<1:29:36,  2.93s/it]
 28%|██▊       | 726/2562 [37:00<1:28:51,  2.90s/it]
 28%|██▊       | 727/2562 [37:03<1:30:04,  2.95s/it]
 28%|██▊       | 728/2562 [37:06<1:27:58,  2.88s/it]
 28%|██▊       | 729/2562 [37:09<1:28:57,  2.91s/it]
 28%|██▊       | 730/2562 [37:12<1:29:42,  2.94s/it]
                                                    
{'loss': 1.0592, 'grad_norm': 5.196182727813721, 'learning_rate': 1.7136952852211274e-05, 'epoch': 0.85}

 28%|██▊       | 730/2562 [37:12<1:29:42,  2.94s/it]
 29%|██▊       | 731/2562 [37:15<1:31:05,  2.99s/it]
 29%|██▊       | 732/2562 [37:17<1:28:41,  2.91s/it]
 29%|██▊       | 733/2562 [37:20<1:29:43,  2.94s/it]
 29%|██▊       | 734/2562 [37:23<1:28:28,  2.90s/it]
 29%|██▊       | 735/2562 [37:26<1:29:02,  2.92s/it]
 29%|██▊       | 736/2562 [37:29<1:29:14,  2.93s/it]
 29%|██▉       | 737/2562 [37:32<1:28:43,  2.92s/it]
 29%|██▉       | 738/2562 [37:35<1:29:34,  2.95s/it]
 29%|██▉       | 739/2562 [37:38<1:29:01,  2.93s/it]
 29%|██▉       | 740/2562 [37:41<1:28:26,  2.91s/it]
                                                    
{'loss': 1.0731, 'grad_norm': 6.055704116821289, 'learning_rate': 1.7045914513874815e-05, 'epoch': 0.87}

 29%|██▉       | 740/2562 [37:41<1:28:26,  2.91s/it]
 29%|██▉       | 741/2562 [37:44<1:28:35,  2.92s/it]
 29%|██▉       | 742/2562 [37:47<1:29:16,  2.94s/it]
 29%|██▉       | 743/2562 [37:50<1:28:58,  2.93s/it]
 29%|██▉       | 744/2562 [37:53<1:29:16,  2.95s/it]
 29%|██▉       | 745/2562 [37:56<1:28:32,  2.92s/it]
 29%|██▉       | 746/2562 [37:58<1:27:18,  2.88s/it]
 29%|██▉       | 747/2562 [38:01<1:28:53,  2.94s/it]
 29%|██▉       | 748/2562 [38:04<1:28:41,  2.93s/it]
 29%|██▉       | 749/2562 [38:07<1:28:03,  2.91s/it]
 29%|██▉       | 750/2562 [38:10<1:28:57,  2.95s/it]
                                                    
{'loss': 1.0398, 'grad_norm': 5.10609245300293, 'learning_rate': 1.695370142172614e-05, 'epoch': 0.88}

 29%|██▉       | 750/2562 [38:10<1:28:57,  2.95s/it]
 29%|██▉       | 751/2562 [38:13<1:27:25,  2.90s/it]
 29%|██▉       | 752/2562 [38:16<1:27:32,  2.90s/it]
 29%|██▉       | 753/2562 [38:19<1:28:59,  2.95s/it]
 29%|██▉       | 754/2562 [38:22<1:27:44,  2.91s/it]
 29%|██▉       | 755/2562 [38:25<1:26:32,  2.87s/it]
 30%|██▉       | 756/2562 [38:28<1:28:44,  2.95s/it]
 30%|██▉       | 757/2562 [38:31<1:27:10,  2.90s/it]
 30%|██▉       | 758/2562 [38:33<1:27:27,  2.91s/it]
 30%|██▉       | 759/2562 [38:36<1:27:46,  2.92s/it]
 30%|██▉       | 760/2562 [38:39<1:26:47,  2.89s/it]
                                                    
{'loss': 1.0433, 'grad_norm': 5.292675018310547, 'learning_rate': 1.6860328950303392e-05, 'epoch': 0.89}

 30%|██▉       | 760/2562 [38:39<1:26:47,  2.89s/it]
 30%|██▉       | 761/2562 [38:42<1:27:11,  2.90s/it]
 30%|██▉       | 762/2562 [38:45<1:28:38,  2.95s/it]
 30%|██▉       | 763/2562 [38:48<1:27:33,  2.92s/it]
 30%|██▉       | 764/2562 [38:51<1:26:07,  2.87s/it]
 30%|██▉       | 765/2562 [38:54<1:28:15,  2.95s/it]
 30%|██▉       | 766/2562 [38:57<1:26:47,  2.90s/it]
 30%|██▉       | 767/2562 [39:00<1:26:37,  2.90s/it]
 30%|██▉       | 768/2562 [39:03<1:27:33,  2.93s/it]
 30%|███       | 769/2562 [39:05<1:25:49,  2.87s/it]
 30%|███       | 770/2562 [39:08<1:26:27,  2.89s/it]
                                                    
{'loss': 1.0475, 'grad_norm': 5.287570476531982, 'learning_rate': 1.676581266744615e-05, 'epoch': 0.9}

 30%|███       | 770/2562 [39:08<1:26:27,  2.89s/it]
 30%|███       | 771/2562 [39:11<1:27:27,  2.93s/it]
 30%|███       | 772/2562 [39:14<1:26:55,  2.91s/it]
 30%|███       | 773/2562 [39:17<1:26:48,  2.91s/it]
 30%|███       | 774/2562 [39:20<1:27:16,  2.93s/it]
 30%|███       | 775/2562 [39:23<1:26:11,  2.89s/it]
 30%|███       | 776/2562 [39:26<1:26:49,  2.92s/it]
 30%|███       | 777/2562 [39:29<1:26:46,  2.92s/it]
 30%|███       | 778/2562 [39:32<1:25:52,  2.89s/it]
 30%|███       | 779/2562 [39:34<1:25:48,  2.89s/it]
 30%|███       | 780/2562 [39:38<1:27:26,  2.94s/it]
                                                    
{'loss': 1.0444, 'grad_norm': 4.72074556350708, 'learning_rate': 1.667016833169979e-05, 'epoch': 0.91}

 30%|███       | 780/2562 [39:38<1:27:26,  2.94s/it]
 30%|███       | 781/2562 [39:40<1:25:50,  2.89s/it]
 31%|███       | 782/2562 [39:44<1:28:22,  2.98s/it]
 31%|███       | 783/2562 [39:46<1:27:08,  2.94s/it]
 31%|███       | 784/2562 [39:49<1:27:08,  2.94s/it]
 31%|███       | 785/2562 [39:52<1:27:09,  2.94s/it]
 31%|███       | 786/2562 [39:55<1:27:32,  2.96s/it]
 31%|███       | 787/2562 [39:58<1:26:29,  2.92s/it]
 31%|███       | 788/2562 [40:01<1:28:32,  2.99s/it]
 31%|███       | 789/2562 [40:04<1:26:34,  2.93s/it]
 31%|███       | 790/2562 [40:07<1:26:07,  2.92s/it]
                                                    
{'loss': 1.0271, 'grad_norm': 5.356393337249756, 'learning_rate': 1.657341188968811e-05, 'epoch': 0.92}

 31%|███       | 790/2562 [40:07<1:26:07,  2.92s/it]
 31%|███       | 791/2562 [40:10<1:26:37,  2.93s/it]
 31%|███       | 792/2562 [40:13<1:27:49,  2.98s/it]
 31%|███       | 793/2562 [40:16<1:26:30,  2.93s/it]
 31%|███       | 794/2562 [40:19<1:26:35,  2.94s/it]
 31%|███       | 795/2562 [40:22<1:26:14,  2.93s/it]
 31%|███       | 796/2562 [40:24<1:25:24,  2.90s/it]
 31%|███       | 797/2562 [40:27<1:26:05,  2.93s/it]
 31%|███       | 798/2562 [40:30<1:24:38,  2.88s/it]
 31%|███       | 799/2562 [40:33<1:23:49,  2.85s/it]
 31%|███       | 800/2562 [40:36<1:25:38,  2.92s/it]
                                                    
{'loss': 1.064, 'grad_norm': 5.321006774902344, 'learning_rate': 1.6475559473454558e-05, 'epoch': 0.94}

 31%|███       | 800/2562 [40:36<1:25:38,  2.92s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.62it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.82it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.27it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.92it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.56it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.75it/s]

 10%|█         | 14/140 [00:01<00:12,  9.77it/s]

 11%|█         | 15/140 [00:01<00:12,  9.82it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.82it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.84it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.61it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.56it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.52it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.46it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.57it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.42it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.12it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.17it/s]

 20%|██        | 28/140 [00:02<00:11,  9.47it/s]

 21%|██        | 29/140 [00:02<00:11,  9.32it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.01it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.01it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.24it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.71it/s]

 25%|██▌       | 35/140 [00:03<00:10,  9.77it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.81it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.85it/s]

 28%|██▊       | 39/140 [00:03<00:10,  9.96it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.95it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.95it/s]

 31%|███       | 43/140 [00:04<00:09,  9.99it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.95it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.91it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.66it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.31it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.25it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.16it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.21it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.39it/s]

 37%|███▋      | 52/140 [00:05<00:09,  9.55it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.35it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.39it/s]

 40%|████      | 56/140 [00:05<00:08,  9.72it/s]

 41%|████      | 57/140 [00:05<00:08,  9.50it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.39it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.49it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.57it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.79it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.61it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.56it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.52it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.50it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.72it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.81it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.89it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.89it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.88it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.64it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.26it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.39it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.30it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.20it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.09it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.11it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.31it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.17it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.16it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.08it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.99it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.06it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.26it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.44it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.28it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.04it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.13it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.37it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.25it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.30it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.48it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.27it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.36it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.53it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.75it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.63it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.52it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.60it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.65it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.83it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.82it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.52it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.55it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.31it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.42it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.12it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.09it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.26it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.24it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.33it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.13it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.92it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.68it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.76it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.04it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.50it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.59it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.60it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.44it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.48it/s]

100%|██████████| 140/140 [00:14<00:00,  8.09it/s]
                                                    

                                                 
{'eval_loss': 1.0804295539855957, 'eval_runtime': 15.1242, 'eval_samples_per_second': 18.513, 'eval_steps_per_second': 9.257, 'epoch': 0.94}

 31%|███       | 800/2562 [40:51<1:25:38,  2.92s/it]

100%|██████████| 140/140 [00:14<00:00,  8.09it/s]

                                                 
 31%|███▏      | 801/2562 [40:54<3:38:35,  7.45s/it]
 31%|███▏      | 802/2562 [40:57<2:57:52,  6.06s/it]
 31%|███▏      | 803/2562 [41:00<2:31:29,  5.17s/it]
 31%|███▏      | 804/2562 [41:03<2:11:47,  4.50s/it]
 31%|███▏      | 805/2562 [41:06<1:58:47,  4.06s/it]
 31%|███▏      | 806/2562 [41:09<1:49:58,  3.76s/it]
 31%|███▏      | 807/2562 [41:12<1:41:32,  3.47s/it]
 32%|███▏      | 808/2562 [41:15<1:36:43,  3.31s/it]
 32%|███▏      | 809/2562 [41:18<1:34:19,  3.23s/it]
 32%|███▏      | 810/2562 [41:21<1:30:56,  3.11s/it]
                                                    
{'loss': 1.0427, 'grad_norm': 5.08799409866333, 'learning_rate': 1.6376627397772576e-05, 'epoch': 0.95}

 32%|███▏      | 810/2562 [41:21<1:30:56,  3.11s/it]
 32%|███▏      | 811/2562 [41:23<1:27:28,  3.00s/it]
 32%|███▏      | 812/2562 [41:26<1:27:21,  3.00s/it]
 32%|███▏      | 813/2562 [41:29<1:26:14,  2.96s/it]
 32%|███▏      | 814/2562 [41:32<1:26:11,  2.96s/it]
 32%|███▏      | 815/2562 [41:35<1:26:18,  2.96s/it]
 32%|███▏      | 816/2562 [41:38<1:26:26,  2.97s/it]
 32%|███▏      | 817/2562 [41:41<1:24:45,  2.91s/it]
 32%|███▏      | 818/2562 [41:44<1:26:49,  2.99s/it]
 32%|███▏      | 819/2562 [41:47<1:24:27,  2.91s/it]
 32%|███▏      | 820/2562 [41:50<1:24:14,  2.90s/it]
                                                    
{'loss': 1.0158, 'grad_norm': 5.187337398529053, 'learning_rate': 1.6276632157425475e-05, 'epoch': 0.96}

 32%|███▏      | 820/2562 [41:50<1:24:14,  2.90s/it]
 32%|███▏      | 821/2562 [41:53<1:25:28,  2.95s/it]
 32%|███▏      | 822/2562 [41:56<1:24:33,  2.92s/it]
 32%|███▏      | 823/2562 [41:59<1:25:39,  2.96s/it]
 32%|███▏      | 824/2562 [42:02<1:26:21,  2.98s/it]
 32%|███▏      | 825/2562 [42:05<1:25:08,  2.94s/it]
 32%|███▏      | 826/2562 [42:08<1:25:44,  2.96s/it]
 32%|███▏      | 827/2562 [42:10<1:25:06,  2.94s/it]
 32%|███▏      | 828/2562 [42:13<1:24:39,  2.93s/it]
 32%|███▏      | 829/2562 [42:17<1:26:25,  2.99s/it]
 32%|███▏      | 830/2562 [42:19<1:24:48,  2.94s/it]
                                                    
{'loss': 0.9972, 'grad_norm': 4.8135528564453125, 'learning_rate': 1.617559042445625e-05, 'epoch': 0.97}

 32%|███▏      | 830/2562 [42:19<1:24:48,  2.94s/it]
 32%|███▏      | 831/2562 [42:22<1:24:34,  2.93s/it]
 32%|███▏      | 832/2562 [42:25<1:24:13,  2.92s/it]
 33%|███▎      | 833/2562 [42:28<1:24:20,  2.93s/it]
 33%|███▎      | 834/2562 [42:31<1:24:44,  2.94s/it]
 33%|███▎      | 835/2562 [42:34<1:24:50,  2.95s/it]
 33%|███▎      | 836/2562 [42:37<1:24:12,  2.93s/it]
 33%|███▎      | 837/2562 [42:40<1:22:55,  2.88s/it]
 33%|███▎      | 838/2562 [42:43<1:24:23,  2.94s/it]
 33%|███▎      | 839/2562 [42:46<1:24:11,  2.93s/it]
 33%|███▎      | 840/2562 [42:49<1:25:24,  2.98s/it]
                                                    
{'loss': 1.0726, 'grad_norm': 5.075769424438477, 'learning_rate': 1.607351904538792e-05, 'epoch': 0.98}

 33%|███▎      | 840/2562 [42:49<1:25:24,  2.98s/it]
 33%|███▎      | 841/2562 [42:52<1:25:47,  2.99s/it]
 33%|███▎      | 842/2562 [42:55<1:24:20,  2.94s/it]
 33%|███▎      | 843/2562 [42:58<1:24:14,  2.94s/it]
 33%|███▎      | 844/2562 [43:01<1:25:11,  2.98s/it]
 33%|███▎      | 845/2562 [43:03<1:23:21,  2.91s/it]
 33%|███▎      | 846/2562 [43:06<1:22:57,  2.90s/it]
 33%|███▎      | 847/2562 [43:09<1:22:43,  2.89s/it]
 33%|███▎      | 848/2562 [43:12<1:23:47,  2.93s/it]
 33%|███▎      | 849/2562 [43:15<1:22:25,  2.89s/it]
 33%|███▎      | 850/2562 [43:18<1:24:00,  2.94s/it]
                                                    
{'loss': 1.0483, 'grad_norm': 5.296349048614502, 'learning_rate': 1.597043503841471e-05, 'epoch': 0.99}

 33%|███▎      | 850/2562 [43:18<1:24:00,  2.94s/it]
 33%|███▎      | 851/2562 [43:21<1:23:09,  2.92s/it]
 33%|███▎      | 852/2562 [43:24<1:22:40,  2.90s/it]
 33%|███▎      | 853/2562 [43:27<1:22:21,  2.89s/it]
 33%|███▎      | 854/2562 [43:29<1:22:32,  2.90s/it]
 33%|███▎      | 855/2562 [43:35<1:42:17,  3.60s/it]
 33%|███▎      | 856/2562 [43:38<1:36:27,  3.39s/it]
 33%|███▎      | 857/2562 [43:41<1:33:38,  3.30s/it]
 33%|███▎      | 858/2562 [43:44<1:30:43,  3.19s/it]
 34%|███▎      | 859/2562 [43:46<1:27:38,  3.09s/it]
 34%|███▎      | 860/2562 [43:49<1:25:45,  3.02s/it]
                                                    
{'loss': 1.1265, 'grad_norm': 5.170070171356201, 'learning_rate': 1.5866355590564637e-05, 'epoch': 1.01}

 34%|███▎      | 860/2562 [43:49<1:25:45,  3.02s/it]
 34%|███▎      | 861/2562 [43:52<1:26:11,  3.04s/it]
 34%|███▎      | 862/2562 [43:55<1:24:22,  2.98s/it]
 34%|███▎      | 863/2562 [43:58<1:25:14,  3.01s/it]
 34%|███▎      | 864/2562 [44:01<1:25:41,  3.03s/it]
 34%|███▍      | 865/2562 [44:04<1:24:03,  2.97s/it]
 34%|███▍      | 866/2562 [44:07<1:23:44,  2.96s/it]
 34%|███▍      | 867/2562 [44:10<1:23:34,  2.96s/it]
 34%|███▍      | 868/2562 [44:13<1:23:59,  2.97s/it]
 34%|███▍      | 869/2562 [44:16<1:22:54,  2.94s/it]
 34%|███▍      | 870/2562 [44:19<1:23:49,  2.97s/it]
                                                    
{'loss': 0.9757, 'grad_norm': 4.940195560455322, 'learning_rate': 1.5761298054833947e-05, 'epoch': 1.02}

 34%|███▍      | 870/2562 [44:19<1:23:49,  2.97s/it]
 34%|███▍      | 871/2562 [44:22<1:22:56,  2.94s/it]
 34%|███▍      | 872/2562 [44:25<1:21:45,  2.90s/it]
 34%|███▍      | 873/2562 [44:28<1:22:52,  2.94s/it]
 34%|███▍      | 874/2562 [44:31<1:22:02,  2.92s/it]
 34%|███▍      | 875/2562 [44:33<1:20:50,  2.88s/it]
 34%|███▍      | 876/2562 [44:37<1:22:37,  2.94s/it]
 34%|███▍      | 877/2562 [44:39<1:22:16,  2.93s/it]
 34%|███▍      | 878/2562 [44:42<1:22:47,  2.95s/it]
 34%|███▍      | 879/2562 [44:45<1:23:33,  2.98s/it]
 34%|███▍      | 880/2562 [44:48<1:21:43,  2.92s/it]
                                                    
{'loss': 0.9664, 'grad_norm': 5.029247760772705, 'learning_rate': 1.565527994729389e-05, 'epoch': 1.03}

 34%|███▍      | 880/2562 [44:48<1:21:43,  2.92s/it]
 34%|███▍      | 881/2562 [44:51<1:20:54,  2.89s/it]
 34%|███▍      | 882/2562 [44:54<1:22:51,  2.96s/it]
 34%|███▍      | 883/2562 [44:57<1:21:51,  2.92s/it]
 35%|███▍      | 884/2562 [45:00<1:21:28,  2.91s/it]
 35%|███▍      | 885/2562 [45:03<1:21:58,  2.93s/it]
 35%|███▍      | 886/2562 [45:06<1:21:28,  2.92s/it]
 35%|███▍      | 887/2562 [45:09<1:21:13,  2.91s/it]
 35%|███▍      | 888/2562 [45:12<1:24:11,  3.02s/it]
 35%|███▍      | 889/2562 [45:15<1:22:41,  2.97s/it]
 35%|███▍      | 890/2562 [45:18<1:22:53,  2.97s/it]
                                                    
{'loss': 0.9394, 'grad_norm': 5.133368968963623, 'learning_rate': 1.5548318944170276e-05, 'epoch': 1.04}

 35%|███▍      | 890/2562 [45:18<1:22:53,  2.97s/it]
 35%|███▍      | 891/2562 [45:21<1:23:00,  2.98s/it]
 35%|███▍      | 892/2562 [45:24<1:22:04,  2.95s/it]
 35%|███▍      | 893/2562 [45:27<1:21:59,  2.95s/it]
 35%|███▍      | 894/2562 [45:30<1:23:06,  2.99s/it]
 35%|███▍      | 895/2562 [45:33<1:21:35,  2.94s/it]
 35%|███▍      | 896/2562 [45:35<1:20:37,  2.90s/it]
 35%|███▌      | 897/2562 [45:38<1:21:26,  2.94s/it]
 35%|███▌      | 898/2562 [45:41<1:20:37,  2.91s/it]
 35%|███▌      | 899/2562 [45:44<1:21:01,  2.92s/it]
 35%|███▌      | 900/2562 [45:47<1:21:53,  2.96s/it]
                                                    
{'loss': 0.957, 'grad_norm': 4.934557914733887, 'learning_rate': 1.544043287889635e-05, 'epoch': 1.05}

 35%|███▌      | 900/2562 [45:47<1:21:53,  2.96s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.65it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.68it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.17it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.85it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.53it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.77it/s]

 11%|█         | 15/140 [00:01<00:12,  9.82it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.81it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.85it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.59it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.52it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.49it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.44it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.53it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.39it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.09it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.16it/s]

 20%|██        | 28/140 [00:02<00:11,  9.51it/s]

 21%|██        | 29/140 [00:02<00:11,  9.37it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.07it/s]

 22%|██▏       | 31/140 [00:03<00:11,  9.08it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.32it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.73it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.85it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.87it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.90it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.96it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.93it/s]

 30%|███       | 42/140 [00:04<00:09,  9.92it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.91it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.86it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.62it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.29it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.28it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.19it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.26it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.45it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.46it/s]

 40%|████      | 56/140 [00:05<00:08,  9.74it/s]

 41%|████      | 57/140 [00:05<00:08,  9.51it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.38it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.57it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.79it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.61it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.55it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.54it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.51it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.76it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.80it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.83it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.84it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.87it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.86it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.84it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.85it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.58it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.21it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.34it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.28it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.19it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.08it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.10it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.33it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.20it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.18it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.08it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.09it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.96it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.06it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.26it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.44it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.28it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.02it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.13it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.30it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.36it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.51it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.34it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.40it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.71it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.73it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.56it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.46it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.56it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.61it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.60it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.81it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.80it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.50it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.41it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.52it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.32it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.29it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.11it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.03it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.23it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.21it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.38it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.28it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.05it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.82it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.59it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.70it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.02it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.27it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.64it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.67it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.51it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.53it/s]

100%|██████████| 140/140 [00:14<00:00,  7.98it/s]
                                                    

                                                 
{'eval_loss': 1.0557953119277954, 'eval_runtime': 15.1351, 'eval_samples_per_second': 18.5, 'eval_steps_per_second': 9.25, 'epoch': 1.05}

 35%|███▌      | 900/2562 [46:02<1:21:53,  2.96s/it]

100%|██████████| 140/140 [00:14<00:00,  7.98it/s]

                                                 
 35%|███▌      | 901/2562 [46:05<3:26:35,  7.46s/it]
 35%|███▌      | 902/2562 [46:08<2:48:51,  6.10s/it]
 35%|███▌      | 903/2562 [46:11<2:23:13,  5.18s/it]
 35%|███▌      | 904/2562 [46:14<2:03:18,  4.46s/it]
 35%|███▌      | 905/2562 [46:17<1:52:11,  4.06s/it]
 35%|███▌      | 906/2562 [46:20<1:44:03,  3.77s/it]
 35%|███▌      | 907/2562 [46:23<1:36:11,  3.49s/it]
 35%|███▌      | 908/2562 [46:26<1:31:29,  3.32s/it]
 35%|███▌      | 909/2562 [46:29<1:28:22,  3.21s/it]
 36%|███▌      | 910/2562 [46:32<1:25:06,  3.09s/it]
                                                    
{'loss': 0.9646, 'grad_norm': 5.000693321228027, 'learning_rate': 1.5331639739139477e-05, 'epoch': 1.07}

 36%|███▌      | 910/2562 [46:32<1:25:06,  3.09s/it]
 36%|███▌      | 911/2562 [46:34<1:22:42,  3.01s/it]
 36%|███▌      | 912/2562 [46:37<1:22:41,  3.01s/it]
 36%|███▌      | 913/2562 [46:40<1:21:13,  2.96s/it]
 36%|███▌      | 914/2562 [46:43<1:22:18,  3.00s/it]
 36%|███▌      | 915/2562 [46:46<1:20:52,  2.95s/it]
 36%|███▌      | 916/2562 [46:49<1:20:05,  2.92s/it]
 36%|███▌      | 917/2562 [46:52<1:21:12,  2.96s/it]
 36%|███▌      | 918/2562 [46:55<1:20:10,  2.93s/it]
 36%|███▌      | 919/2562 [46:58<1:20:05,  2.92s/it]
 36%|███▌      | 920/2562 [47:01<1:18:22,  2.86s/it]
                                                    
{'loss': 1.0029, 'grad_norm': 5.345034599304199, 'learning_rate': 1.5221957663802043e-05, 'epoch': 1.08}

 36%|███▌      | 920/2562 [47:01<1:18:22,  2.86s/it]
 36%|███▌      | 921/2562 [47:04<1:19:51,  2.92s/it]
 36%|███▌      | 922/2562 [47:07<1:20:10,  2.93s/it]
 36%|███▌      | 923/2562 [47:09<1:18:56,  2.89s/it]
 36%|███▌      | 924/2562 [47:12<1:20:13,  2.94s/it]
 36%|███▌      | 925/2562 [47:15<1:18:48,  2.89s/it]
 36%|███▌      | 926/2562 [47:18<1:19:58,  2.93s/it]
 36%|███▌      | 927/2562 [47:21<1:19:01,  2.90s/it]
 36%|███▌      | 928/2562 [47:24<1:19:11,  2.91s/it]
 36%|███▋      | 929/2562 [47:27<1:18:53,  2.90s/it]
 36%|███▋      | 930/2562 [47:30<1:18:57,  2.90s/it]
                                                    
{'loss': 0.982, 'grad_norm': 5.138866424560547, 'learning_rate': 1.5111404939997227e-05, 'epoch': 1.09}

 36%|███▋      | 930/2562 [47:30<1:18:57,  2.90s/it]
 36%|███▋      | 931/2562 [47:33<1:18:26,  2.89s/it]
 36%|███▋      | 932/2562 [47:36<1:19:07,  2.91s/it]
 36%|███▋      | 933/2562 [47:39<1:20:39,  2.97s/it]
 36%|███▋      | 934/2562 [47:42<1:20:36,  2.97s/it]
 36%|███▋      | 935/2562 [47:45<1:21:27,  3.00s/it]
 37%|███▋      | 936/2562 [47:48<1:20:44,  2.98s/it]
 37%|███▋      | 937/2562 [47:51<1:20:23,  2.97s/it]
 37%|███▋      | 938/2562 [47:54<1:20:00,  2.96s/it]
 37%|███▋      | 939/2562 [47:56<1:18:56,  2.92s/it]
 37%|███▋      | 940/2562 [47:59<1:17:18,  2.86s/it]
                                                    
{'loss': 0.9937, 'grad_norm': 5.084903240203857, 'learning_rate': 1.5000000000000002e-05, 'epoch': 1.1}

 37%|███▋      | 940/2562 [47:59<1:17:18,  2.86s/it]
 37%|███▋      | 941/2562 [48:02<1:18:21,  2.90s/it]
 37%|███▋      | 942/2562 [48:05<1:17:43,  2.88s/it]
 37%|███▋      | 943/2562 [48:08<1:16:53,  2.85s/it]
 37%|███▋      | 944/2562 [48:11<1:17:39,  2.88s/it]
 37%|███▋      | 945/2562 [48:14<1:17:49,  2.89s/it]
 37%|███▋      | 946/2562 [48:17<1:17:55,  2.89s/it]
 37%|███▋      | 947/2562 [48:19<1:17:42,  2.89s/it]
 37%|███▋      | 948/2562 [48:22<1:18:50,  2.93s/it]
 37%|███▋      | 949/2562 [48:25<1:19:11,  2.95s/it]
 37%|███▋      | 950/2562 [48:28<1:17:43,  2.89s/it]
                                                    
{'loss': 0.9783, 'grad_norm': 5.721775054931641, 'learning_rate': 1.4887761418173947e-05, 'epoch': 1.11}

 37%|███▋      | 950/2562 [48:28<1:17:43,  2.89s/it]
 37%|███▋      | 951/2562 [48:31<1:17:33,  2.89s/it]
 37%|███▋      | 952/2562 [48:34<1:18:22,  2.92s/it]
 37%|███▋      | 953/2562 [48:37<1:17:00,  2.87s/it]
 37%|███▋      | 954/2562 [48:40<1:16:44,  2.86s/it]
 37%|███▋      | 955/2562 [48:43<1:18:08,  2.92s/it]
 37%|███▋      | 956/2562 [48:45<1:16:51,  2.87s/it]
 37%|███▋      | 957/2562 [48:48<1:15:50,  2.84s/it]
 37%|███▋      | 958/2562 [48:51<1:16:39,  2.87s/it]
 37%|███▋      | 959/2562 [48:54<1:16:42,  2.87s/it]
 37%|███▋      | 960/2562 [48:57<1:16:17,  2.86s/it]
                                                    
{'loss': 0.98, 'grad_norm': 4.778964519500732, 'learning_rate': 1.4774707907874392e-05, 'epoch': 1.12}

 37%|███▋      | 960/2562 [48:57<1:16:17,  2.86s/it]
 38%|███▊      | 961/2562 [49:00<1:17:31,  2.91s/it]
 38%|███▊      | 962/2562 [49:03<1:16:56,  2.89s/it]
 38%|███▊      | 963/2562 [49:05<1:16:02,  2.85s/it]
 38%|███▊      | 964/2562 [49:08<1:16:25,  2.87s/it]
 38%|███▊      | 965/2562 [49:11<1:17:13,  2.90s/it]
 38%|███▊      | 966/2562 [49:14<1:17:29,  2.91s/it]
 38%|███▊      | 967/2562 [49:17<1:18:29,  2.95s/it]
 38%|███▊      | 968/2562 [49:20<1:17:01,  2.90s/it]
 38%|███▊      | 969/2562 [49:23<1:15:57,  2.86s/it]
 38%|███▊      | 970/2562 [49:26<1:17:37,  2.93s/it]
                                                    
{'loss': 1.0004, 'grad_norm': 4.717801570892334, 'learning_rate': 1.4660858318328348e-05, 'epoch': 1.14}

 38%|███▊      | 970/2562 [49:26<1:17:37,  2.93s/it]
 38%|███▊      | 971/2562 [49:29<1:17:50,  2.94s/it]
 38%|███▊      | 972/2562 [49:32<1:17:46,  2.93s/it]
 38%|███▊      | 973/2562 [49:35<1:17:58,  2.94s/it]
 38%|███▊      | 974/2562 [49:38<1:16:56,  2.91s/it]
 38%|███▊      | 975/2562 [49:41<1:16:47,  2.90s/it]
 38%|███▊      | 976/2562 [49:44<1:17:15,  2.92s/it]
 38%|███▊      | 977/2562 [49:47<1:18:13,  2.96s/it]
 38%|███▊      | 978/2562 [49:49<1:17:44,  2.94s/it]
 38%|███▊      | 979/2562 [49:52<1:17:59,  2.96s/it]
 38%|███▊      | 980/2562 [49:55<1:17:33,  2.94s/it]
                                                    
{'loss': 0.9442, 'grad_norm': 4.790687561035156, 'learning_rate': 1.4546231631491827e-05, 'epoch': 1.15}

 38%|███▊      | 980/2562 [49:55<1:17:33,  2.94s/it]
 38%|███▊      | 981/2562 [49:58<1:16:11,  2.89s/it]
 38%|███▊      | 982/2562 [50:01<1:17:23,  2.94s/it]
 38%|███▊      | 983/2562 [50:04<1:16:47,  2.92s/it]
 38%|███▊      | 984/2562 [50:07<1:17:51,  2.96s/it]
 38%|███▊      | 985/2562 [50:10<1:18:06,  2.97s/it]
 38%|███▊      | 986/2562 [50:13<1:16:59,  2.93s/it]
 39%|███▊      | 987/2562 [50:16<1:15:16,  2.87s/it]
 39%|███▊      | 988/2562 [50:19<1:15:50,  2.89s/it]
 39%|███▊      | 989/2562 [50:21<1:15:18,  2.87s/it]
 39%|███▊      | 990/2562 [50:24<1:15:04,  2.87s/it]
                                                    
{'loss': 0.9794, 'grad_norm': 4.972203731536865, 'learning_rate': 1.4430846958884995e-05, 'epoch': 1.16}

 39%|███▊      | 990/2562 [50:24<1:15:04,  2.87s/it]
 39%|███▊      | 991/2562 [50:27<1:17:28,  2.96s/it]
 39%|███▊      | 992/2562 [50:30<1:17:01,  2.94s/it]
 39%|███▉      | 993/2562 [50:33<1:17:19,  2.96s/it]
 39%|███▉      | 994/2562 [50:36<1:17:18,  2.96s/it]
 39%|███▉      | 995/2562 [50:39<1:15:26,  2.89s/it]
 39%|███▉      | 996/2562 [50:42<1:16:43,  2.94s/it]
 39%|███▉      | 997/2562 [50:45<1:17:22,  2.97s/it]
 39%|███▉      | 998/2562 [50:48<1:16:39,  2.94s/it]
 39%|███▉      | 999/2562 [50:51<1:16:30,  2.94s/it]
 39%|███▉      | 1000/2562 [50:54<1:15:15,  2.89s/it]
                                                     
{'loss': 0.9838, 'grad_norm': 5.111540794372559, 'learning_rate': 1.4314723538405752e-05, 'epoch': 1.17}

 39%|███▉      | 1000/2562 [50:54<1:15:15,  2.89s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.50it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.62it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.14it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.79it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.47it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.68it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.71it/s]

 10%|█         | 14/140 [00:01<00:12,  9.75it/s]

 11%|█         | 15/140 [00:01<00:12,  9.79it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.78it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.82it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.59it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.53it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.65it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.40it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.37it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.51it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.36it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.05it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.11it/s]

 20%|██        | 28/140 [00:02<00:11,  9.44it/s]

 21%|██        | 29/140 [00:02<00:11,  9.27it/s]

 21%|██▏       | 30/140 [00:03<00:12,  8.97it/s]

 22%|██▏       | 31/140 [00:03<00:12,  8.97it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.19it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.64it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.79it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.81it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.86it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.93it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.91it/s]

 31%|███       | 43/140 [00:04<00:09,  9.94it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.92it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.89it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.63it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.29it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.27it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.13it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.23it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.42it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.42it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.39it/s]

 40%|████      | 56/140 [00:05<00:08,  9.65it/s]

 41%|████      | 57/140 [00:05<00:08,  9.45it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.36it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.45it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.56it/s]

 44%|████▎     | 61/140 [00:06<00:08,  9.68it/s]

 44%|████▍     | 62/140 [00:06<00:08,  9.75it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.54it/s]

 46%|████▌     | 64/140 [00:06<00:08,  9.49it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.49it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.46it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.68it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.74it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.78it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.79it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.84it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.85it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.86it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.81it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.57it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.17it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.31it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.27it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.18it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.10it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.10it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.29it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.18it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.16it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.05it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.07it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.96it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.05it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.26it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.42it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.25it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.01it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.10it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.34it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.22it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.29it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.47it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.28it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.36it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.54it/s]

 75%|███████▌  | 105/140 [00:11<00:03,  9.75it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.61it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.51it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.59it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.65it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.81it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.80it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.51it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.41it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.54it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.40it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.10it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.07it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.22it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.38it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.27it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.05it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.81it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.56it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.68it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  9.01it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.49it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.62it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.48it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.49it/s]

100%|██████████| 140/140 [00:14<00:00,  8.18it/s]
                                                     

                                                 
{'eval_loss': 1.0336856842041016, 'eval_runtime': 15.1734, 'eval_samples_per_second': 18.453, 'eval_steps_per_second': 9.227, 'epoch': 1.17}

 39%|███▉      | 1000/2562 [51:09<1:15:15,  2.89s/it]

100%|██████████| 140/140 [00:14<00:00,  8.18it/s]

                                                 /opt/conda/lib/python3.11/site-packages/peft/utils/save_and_load.py:372: UserWarning: Could not find a config file in  - will assume that the vocabulary was not modified.
  warnings.warn(

 39%|███▉      | 1001/2562 [51:13<3:25:06,  7.88s/it]
 39%|███▉      | 1002/2562 [51:16<2:47:00,  6.42s/it]
 39%|███▉      | 1003/2562 [51:19<2:19:27,  5.37s/it]
 39%|███▉      | 1004/2562 [51:22<1:59:40,  4.61s/it]
 39%|███▉      | 1005/2562 [51:25<1:48:01,  4.16s/it]
 39%|███▉      | 1006/2562 [51:28<1:37:44,  3.77s/it]
 39%|███▉      | 1007/2562 [51:31<1:30:28,  3.49s/it]
 39%|███▉      | 1008/2562 [51:34<1:25:58,  3.32s/it]
 39%|███▉      | 1009/2562 [51:37<1:21:55,  3.17s/it]
 39%|███▉      | 1010/2562 [51:39<1:18:51,  3.05s/it]
                                                     
{'loss': 0.9503, 'grad_norm': 4.875945568084717, 'learning_rate': 1.4197880731122221e-05, 'epoch': 1.18}

 39%|███▉      | 1010/2562 [51:39<1:18:51,  3.05s/it]
 39%|███▉      | 1011/2562 [51:42<1:17:42,  3.01s/it]
 40%|███▉      | 1012/2562 [51:45<1:15:36,  2.93s/it]
 40%|███▉      | 1013/2562 [51:48<1:14:46,  2.90s/it]
 40%|███▉      | 1014/2562 [51:51<1:15:52,  2.94s/it]
 40%|███▉      | 1015/2562 [51:54<1:14:41,  2.90s/it]
 40%|███▉      | 1016/2562 [51:57<1:14:31,  2.89s/it]
 40%|███▉      | 1017/2562 [52:00<1:15:50,  2.95s/it]
 40%|███▉      | 1018/2562 [52:02<1:14:39,  2.90s/it]
 40%|███▉      | 1019/2562 [52:05<1:14:06,  2.88s/it]
 40%|███▉      | 1020/2562 [52:08<1:14:17,  2.89s/it]
                                                     
{'loss': 0.9559, 'grad_norm': 4.93674898147583, 'learning_rate': 1.4080338018044712e-05, 'epoch': 1.19}

 40%|███▉      | 1020/2562 [52:08<1:14:17,  2.89s/it]
 40%|███▉      | 1021/2562 [52:11<1:13:45,  2.87s/it]
 40%|███▉      | 1022/2562 [52:14<1:14:12,  2.89s/it]
 40%|███▉      | 1023/2562 [52:17<1:15:03,  2.93s/it]
 40%|███▉      | 1024/2562 [52:20<1:14:12,  2.90s/it]
 40%|████      | 1025/2562 [52:23<1:13:07,  2.85s/it]
 40%|████      | 1026/2562 [52:26<1:14:27,  2.91s/it]
 40%|████      | 1027/2562 [52:28<1:14:42,  2.92s/it]
 40%|████      | 1028/2562 [52:31<1:14:56,  2.93s/it]
 40%|████      | 1029/2562 [52:34<1:15:35,  2.96s/it]
 40%|████      | 1030/2562 [52:37<1:15:36,  2.96s/it]
                                                     
{'loss': 0.9655, 'grad_norm': 5.131695747375488, 'learning_rate': 1.3962114996877685e-05, 'epoch': 1.21}

 40%|████      | 1030/2562 [52:37<1:15:36,  2.96s/it]
 40%|████      | 1031/2562 [52:40<1:14:39,  2.93s/it]
 40%|████      | 1032/2562 [52:43<1:15:21,  2.96s/it]
 40%|████      | 1033/2562 [52:46<1:13:58,  2.90s/it]
 40%|████      | 1034/2562 [52:49<1:13:49,  2.90s/it]
 40%|████      | 1035/2562 [52:52<1:15:14,  2.96s/it]
 40%|████      | 1036/2562 [52:55<1:14:24,  2.93s/it]
 40%|████      | 1037/2562 [52:58<1:13:17,  2.88s/it]
 41%|████      | 1038/2562 [53:01<1:13:56,  2.91s/it]
 41%|████      | 1039/2562 [53:03<1:12:18,  2.85s/it]
 41%|████      | 1040/2562 [53:06<1:12:43,  2.87s/it]
                                                     
{'loss': 0.9692, 'grad_norm': 5.361415863037109, 'learning_rate': 1.3843231378752252e-05, 'epoch': 1.22}

 41%|████      | 1040/2562 [53:06<1:12:43,  2.87s/it]
 41%|████      | 1041/2562 [53:09<1:12:45,  2.87s/it]
 41%|████      | 1042/2562 [53:12<1:13:44,  2.91s/it]
 41%|████      | 1043/2562 [53:15<1:14:02,  2.92s/it]
 41%|████      | 1044/2562 [53:18<1:14:21,  2.94s/it]
 41%|████      | 1045/2562 [53:21<1:13:06,  2.89s/it]
 41%|████      | 1046/2562 [53:24<1:13:52,  2.92s/it]
 41%|████      | 1047/2562 [53:27<1:14:29,  2.95s/it]
 41%|████      | 1048/2562 [53:30<1:12:41,  2.88s/it]
 41%|████      | 1049/2562 [53:33<1:12:54,  2.89s/it]
 41%|████      | 1050/2562 [53:35<1:13:13,  2.91s/it]
                                                     
{'loss': 0.9645, 'grad_norm': 5.1561174392700195, 'learning_rate': 1.3723706984939783e-05, 'epoch': 1.23}

 41%|████      | 1050/2562 [53:35<1:13:13,  2.91s/it]
 41%|████      | 1051/2562 [53:38<1:13:13,  2.91s/it]
 41%|████      | 1052/2562 [53:41<1:13:34,  2.92s/it]
 41%|████      | 1053/2562 [53:44<1:12:33,  2.88s/it]
 41%|████      | 1054/2562 [53:47<1:12:49,  2.90s/it]
 41%|████      | 1055/2562 [53:50<1:14:43,  2.97s/it]
 41%|████      | 1056/2562 [53:53<1:14:25,  2.97s/it]
 41%|████▏     | 1057/2562 [53:56<1:13:14,  2.92s/it]
 41%|████▏     | 1058/2562 [53:59<1:14:03,  2.95s/it]
 41%|████▏     | 1059/2562 [54:02<1:13:06,  2.92s/it]
 41%|████▏     | 1060/2562 [54:05<1:12:42,  2.90s/it]
                                                     
{'loss': 0.9364, 'grad_norm': 4.781491279602051, 'learning_rate': 1.3603561743547125e-05, 'epoch': 1.24}

 41%|████▏     | 1060/2562 [54:05<1:12:42,  2.90s/it]
 41%|████▏     | 1061/2562 [54:08<1:13:10,  2.92s/it]
 41%|████▏     | 1062/2562 [54:10<1:11:56,  2.88s/it]
 41%|████▏     | 1063/2562 [54:13<1:12:32,  2.90s/it]
 42%|████▏     | 1064/2562 [54:16<1:12:53,  2.92s/it]
 42%|████▏     | 1065/2562 [54:19<1:12:59,  2.93s/it]
 42%|████▏     | 1066/2562 [54:22<1:12:40,  2.91s/it]
 42%|████▏     | 1067/2562 [54:25<1:13:58,  2.97s/it]
 42%|████▏     | 1068/2562 [54:28<1:13:05,  2.94s/it]
 42%|████▏     | 1069/2562 [54:31<1:12:40,  2.92s/it]
 42%|████▏     | 1070/2562 [54:34<1:13:13,  2.95s/it]
                                                     
{'loss': 0.9589, 'grad_norm': 5.274555206298828, 'learning_rate': 1.3482815686194033e-05, 'epoch': 1.25}

 42%|████▏     | 1070/2562 [54:34<1:13:13,  2.95s/it]
 42%|████▏     | 1071/2562 [54:37<1:12:30,  2.92s/it]
 42%|████▏     | 1072/2562 [54:40<1:12:15,  2.91s/it]
 42%|████▏     | 1073/2562 [54:43<1:13:05,  2.95s/it]
 42%|████▏     | 1074/2562 [54:46<1:12:58,  2.94s/it]
 42%|████▏     | 1075/2562 [54:49<1:12:05,  2.91s/it]
 42%|████▏     | 1076/2562 [54:52<1:12:49,  2.94s/it]
 42%|████▏     | 1077/2562 [54:54<1:11:47,  2.90s/it]
 42%|████▏     | 1078/2562 [54:57<1:11:43,  2.90s/it]
 42%|████▏     | 1079/2562 [55:00<1:11:38,  2.90s/it]
 42%|████▏     | 1080/2562 [55:03<1:10:49,  2.87s/it]
                                                     
{'loss': 0.9576, 'grad_norm': 5.180062770843506, 'learning_rate': 1.3361488944673315e-05, 'epoch': 1.26}

 42%|████▏     | 1080/2562 [55:03<1:10:49,  2.87s/it]
 42%|████▏     | 1081/2562 [55:06<1:11:28,  2.90s/it]
 42%|████▏     | 1082/2562 [55:09<1:10:33,  2.86s/it]
 42%|████▏     | 1083/2562 [55:12<1:10:02,  2.84s/it]
 42%|████▏     | 1084/2562 [55:15<1:11:09,  2.89s/it]
 42%|████▏     | 1085/2562 [55:17<1:11:19,  2.90s/it]
 42%|████▏     | 1086/2562 [55:20<1:10:16,  2.86s/it]
 42%|████▏     | 1087/2562 [55:23<1:12:41,  2.96s/it]
 42%|████▏     | 1088/2562 [55:26<1:12:56,  2.97s/it]
 43%|████▎     | 1089/2562 [55:29<1:11:51,  2.93s/it]
 43%|████▎     | 1090/2562 [55:32<1:12:28,  2.95s/it]
                                                     
{'loss': 0.9558, 'grad_norm': 5.264287948608398, 'learning_rate': 1.3239601747594319e-05, 'epoch': 1.28}

 43%|████▎     | 1090/2562 [55:32<1:12:28,  2.95s/it]
 43%|████▎     | 1091/2562 [55:35<1:12:18,  2.95s/it]
 43%|████▎     | 1092/2562 [55:38<1:12:08,  2.94s/it]
 43%|████▎     | 1093/2562 [55:41<1:12:37,  2.97s/it]
 43%|████▎     | 1094/2562 [55:44<1:11:50,  2.94s/it]
 43%|████▎     | 1095/2562 [55:47<1:12:28,  2.96s/it]
 43%|████▎     | 1096/2562 [55:50<1:13:22,  3.00s/it]
 43%|████▎     | 1097/2562 [55:53<1:12:46,  2.98s/it]
 43%|████▎     | 1098/2562 [55:56<1:11:53,  2.95s/it]
 43%|████▎     | 1099/2562 [55:59<1:12:32,  2.98s/it]
 43%|████▎     | 1100/2562 [56:02<1:11:03,  2.92s/it]
                                                     
{'loss': 0.9209, 'grad_norm': 4.894036293029785, 'learning_rate': 1.3117174417010213e-05, 'epoch': 1.29}

 43%|████▎     | 1100/2562 [56:02<1:11:03,  2.92s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.82it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.76it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.22it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.91it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.57it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

 10%|█         | 14/140 [00:01<00:12,  9.74it/s]

 11%|█         | 15/140 [00:01<00:12,  9.77it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.77it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.80it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.52it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.47it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.61it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.39it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.32it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.45it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.33it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.02it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.12it/s]

 20%|██        | 28/140 [00:02<00:11,  9.48it/s]

 21%|██        | 29/140 [00:02<00:11,  9.33it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.04it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.06it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.30it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.73it/s]

 25%|██▌       | 35/140 [00:03<00:10,  9.78it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.80it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.82it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.93it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.93it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 31%|███       | 43/140 [00:04<00:09,  9.95it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.92it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.88it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.65it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.29it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.26it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.16it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.23it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 37%|███▋      | 52/140 [00:05<00:09,  9.57it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.39it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.41it/s]

 40%|████      | 56/140 [00:05<00:08,  9.74it/s]

 41%|████      | 57/140 [00:05<00:08,  9.48it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.39it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.49it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.59it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.79it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.60it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.53it/s]

 46%|████▋     | 65/140 [00:06<00:07,  9.61it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.40it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.39it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.68it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.75it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.84it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.87it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.87it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.85it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.87it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.61it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.20it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.35it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.28it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.20it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.10it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.13it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.33it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.21it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.18it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.08it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.99it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.06it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.24it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.40it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.24it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.00it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.12it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.29it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.32it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.46it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.26it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.33it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.68it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.71it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.60it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.50it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.59it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.64it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.68it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.86it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.87it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.56it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.44it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.53it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.31it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.11it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.09it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.28it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.25it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.42it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.31it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.08it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.86it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.62it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.71it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  9.02it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.29it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.66it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.69it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.52it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.54it/s]

100%|██████████| 140/140 [00:14<00:00,  8.11it/s]
                                                     

                                                 
{'eval_loss': 1.0162620544433594, 'eval_runtime': 15.1384, 'eval_samples_per_second': 18.496, 'eval_steps_per_second': 9.248, 'epoch': 1.29}

 43%|████▎     | 1100/2562 [56:17<1:11:03,  2.92s/it]

100%|██████████| 140/140 [00:14<00:00,  8.11it/s]

                                                 
 43%|████▎     | 1101/2562 [56:20<3:01:05,  7.44s/it]
 43%|████▎     | 1102/2562 [56:23<2:29:31,  6.14s/it]
 43%|████▎     | 1103/2562 [56:26<2:05:20,  5.15s/it]
 43%|████▎     | 1104/2562 [56:29<1:49:22,  4.50s/it]
 43%|████▎     | 1105/2562 [56:32<1:39:20,  4.09s/it]
 43%|████▎     | 1106/2562 [56:35<1:29:58,  3.71s/it]
 43%|████▎     | 1107/2562 [56:38<1:24:27,  3.48s/it]
 43%|████▎     | 1108/2562 [56:41<1:20:43,  3.33s/it]
 43%|████▎     | 1109/2562 [56:44<1:18:45,  3.25s/it]
 43%|████▎     | 1110/2562 [56:47<1:17:00,  3.18s/it]
                                                     
{'loss': 0.9415, 'grad_norm': 4.639826774597168, 'learning_rate': 1.2994227365029752e-05, 'epoch': 1.3}

 43%|████▎     | 1110/2562 [56:47<1:17:00,  3.18s/it]
 43%|████▎     | 1111/2562 [56:50<1:14:47,  3.09s/it]
 43%|████▎     | 1112/2562 [56:52<1:12:51,  3.01s/it]
 43%|████▎     | 1113/2562 [56:56<1:13:52,  3.06s/it]
 43%|████▎     | 1114/2562 [56:58<1:13:05,  3.03s/it]
 44%|████▎     | 1115/2562 [57:01<1:11:41,  2.97s/it]
 44%|████▎     | 1116/2562 [57:04<1:09:42,  2.89s/it]
 44%|████▎     | 1117/2562 [57:07<1:11:16,  2.96s/it]
 44%|████▎     | 1118/2562 [57:10<1:11:02,  2.95s/it]
 44%|████▎     | 1119/2562 [57:13<1:10:59,  2.95s/it]
 44%|████▎     | 1120/2562 [57:16<1:12:14,  3.01s/it]
                                                     
{'loss': 0.9339, 'grad_norm': 5.026745319366455, 'learning_rate': 1.2870781090413991e-05, 'epoch': 1.31}

 44%|████▎     | 1120/2562 [57:16<1:12:14,  3.01s/it]
 44%|████▍     | 1121/2562 [57:19<1:10:39,  2.94s/it]
 44%|████▍     | 1122/2562 [57:22<1:11:24,  2.98s/it]
 44%|████▍     | 1123/2562 [57:25<1:11:12,  2.97s/it]
 44%|████▍     | 1124/2562 [57:28<1:10:00,  2.92s/it]
 44%|████▍     | 1125/2562 [57:31<1:10:01,  2.92s/it]
 44%|████▍     | 1126/2562 [57:34<1:09:31,  2.90s/it]
 44%|████▍     | 1127/2562 [57:36<1:08:51,  2.88s/it]
 44%|████▍     | 1128/2562 [57:39<1:08:50,  2.88s/it]
 44%|████▍     | 1129/2562 [57:42<1:09:28,  2.91s/it]
 44%|████▍     | 1130/2562 [57:45<1:08:23,  2.87s/it]
                                                     
{'loss': 0.9443, 'grad_norm': 4.955203056335449, 'learning_rate': 1.2746856175158556e-05, 'epoch': 1.32}

 44%|████▍     | 1130/2562 [57:45<1:08:23,  2.87s/it]
 44%|████▍     | 1131/2562 [57:48<1:09:06,  2.90s/it]
 44%|████▍     | 1132/2562 [57:51<1:07:43,  2.84s/it]
 44%|████▍     | 1133/2562 [57:53<1:07:25,  2.83s/it]
 44%|████▍     | 1134/2562 [57:56<1:08:21,  2.87s/it]
 44%|████▍     | 1135/2562 [57:59<1:08:10,  2.87s/it]
 44%|████▍     | 1136/2562 [58:02<1:07:47,  2.85s/it]
 44%|████▍     | 1137/2562 [58:05<1:08:28,  2.88s/it]
 44%|████▍     | 1138/2562 [58:08<1:07:34,  2.85s/it]
 44%|████▍     | 1139/2562 [58:11<1:07:29,  2.85s/it]
 44%|████▍     | 1140/2562 [58:14<1:08:30,  2.89s/it]
                                                     
{'loss': 0.9203, 'grad_norm': 4.5640716552734375, 'learning_rate': 1.2622473281062042e-05, 'epoch': 1.33}

 44%|████▍     | 1140/2562 [58:14<1:08:30,  2.89s/it]
 45%|████▍     | 1141/2562 [58:17<1:08:24,  2.89s/it]
 45%|████▍     | 1142/2562 [58:20<1:09:02,  2.92s/it]
 45%|████▍     | 1143/2562 [58:22<1:08:23,  2.89s/it]
 45%|████▍     | 1144/2562 [58:25<1:07:47,  2.87s/it]
 45%|████▍     | 1145/2562 [58:28<1:09:37,  2.95s/it]
 45%|████▍     | 1146/2562 [58:31<1:09:04,  2.93s/it]
 45%|████▍     | 1147/2562 [58:34<1:08:07,  2.89s/it]
 45%|████▍     | 1148/2562 [58:37<1:09:40,  2.96s/it]
 45%|████▍     | 1149/2562 [58:40<1:08:49,  2.92s/it]
 45%|████▍     | 1150/2562 [58:43<1:08:17,  2.90s/it]
                                                     
{'loss': 0.9134, 'grad_norm': 4.7821736335754395, 'learning_rate': 1.2497653146281113e-05, 'epoch': 1.35}

 45%|████▍     | 1150/2562 [58:43<1:08:17,  2.90s/it]
 45%|████▍     | 1151/2562 [58:46<1:09:07,  2.94s/it]
 45%|████▍     | 1152/2562 [58:49<1:09:02,  2.94s/it]
 45%|████▌     | 1153/2562 [58:52<1:08:39,  2.92s/it]
 45%|████▌     | 1154/2562 [58:55<1:08:58,  2.94s/it]
 45%|████▌     | 1155/2562 [58:58<1:08:55,  2.94s/it]
 45%|████▌     | 1156/2562 [59:00<1:08:37,  2.93s/it]
 45%|████▌     | 1157/2562 [59:03<1:08:15,  2.92s/it]
 45%|████▌     | 1158/2562 [59:06<1:07:17,  2.88s/it]
 45%|████▌     | 1159/2562 [59:09<1:07:42,  2.90s/it]
 45%|████▌     | 1160/2562 [59:12<1:09:06,  2.96s/it]
                                                     
{'loss': 0.9404, 'grad_norm': 5.193197250366211, 'learning_rate': 1.2372416581872857e-05, 'epoch': 1.36}

 45%|████▌     | 1160/2562 [59:12<1:09:06,  2.96s/it]
 45%|████▌     | 1161/2562 [59:15<1:08:33,  2.94s/it]
 45%|████▌     | 1162/2562 [59:18<1:08:13,  2.92s/it]
 45%|████▌     | 1163/2562 [59:21<1:08:45,  2.95s/it]
 45%|████▌     | 1164/2562 [59:24<1:08:04,  2.92s/it]
 45%|████▌     | 1165/2562 [59:27<1:07:52,  2.92s/it]
 46%|████▌     | 1166/2562 [59:30<1:09:46,  3.00s/it]
 46%|████▌     | 1167/2562 [59:33<1:08:51,  2.96s/it]
 46%|████▌     | 1168/2562 [59:36<1:07:36,  2.91s/it]
 46%|████▌     | 1169/2562 [59:39<1:08:30,  2.95s/it]
 46%|████▌     | 1170/2562 [59:42<1:08:16,  2.94s/it]
                                                     
{'loss': 0.9309, 'grad_norm': 5.007082462310791, 'learning_rate': 1.2246784468324993e-05, 'epoch': 1.37}

 46%|████▌     | 1170/2562 [59:42<1:08:16,  2.94s/it]
 46%|████▌     | 1171/2562 [59:45<1:08:36,  2.96s/it]
 46%|████▌     | 1172/2562 [59:47<1:07:17,  2.90s/it]
 46%|████▌     | 1173/2562 [59:50<1:07:05,  2.90s/it]
 46%|████▌     | 1174/2562 [59:53<1:06:59,  2.90s/it]
 46%|████▌     | 1175/2562 [59:56<1:07:30,  2.92s/it]
 46%|████▌     | 1176/2562 [59:59<1:08:17,  2.96s/it]
 46%|████▌     | 1177/2562 [1:00:02<1:08:37,  2.97s/it]
 46%|████▌     | 1178/2562 [1:00:05<1:08:12,  2.96s/it]
 46%|████▌     | 1179/2562 [1:00:08<1:07:32,  2.93s/it]
 46%|████▌     | 1180/2562 [1:00:11<1:08:33,  2.98s/it]
                                                       
{'loss': 0.9474, 'grad_norm': 5.159805774688721, 'learning_rate': 1.2120777752074492e-05, 'epoch': 1.38}

 46%|████▌     | 1180/2562 [1:00:11<1:08:33,  2.98s/it]
 46%|████▌     | 1181/2562 [1:00:14<1:06:48,  2.90s/it]
 46%|████▌     | 1182/2562 [1:00:17<1:05:53,  2.86s/it]
 46%|████▌     | 1183/2562 [1:00:20<1:07:04,  2.92s/it]
 46%|████▌     | 1184/2562 [1:00:23<1:07:15,  2.93s/it]
 46%|████▋     | 1185/2562 [1:00:25<1:06:12,  2.89s/it]
 46%|████▋     | 1186/2562 [1:00:28<1:06:49,  2.91s/it]
 46%|████▋     | 1187/2562 [1:00:31<1:06:11,  2.89s/it]
 46%|████▋     | 1188/2562 [1:00:34<1:05:16,  2.85s/it]
 46%|████▋     | 1189/2562 [1:00:37<1:05:42,  2.87s/it]
 46%|████▋     | 1190/2562 [1:00:40<1:05:15,  2.85s/it]
                                                       
{'loss': 0.9337, 'grad_norm': 4.7778000831604, 'learning_rate': 1.1994417442015243e-05, 'epoch': 1.39}

 46%|████▋     | 1190/2562 [1:00:40<1:05:15,  2.85s/it]
 46%|████▋     | 1191/2562 [1:00:42<1:05:07,  2.85s/it]
 47%|████▋     | 1192/2562 [1:00:45<1:06:04,  2.89s/it]
 47%|████▋     | 1193/2562 [1:00:48<1:05:03,  2.85s/it]
 47%|████▋     | 1194/2562 [1:00:51<1:05:08,  2.86s/it]
 47%|████▋     | 1195/2562 [1:00:54<1:05:36,  2.88s/it]
 47%|████▋     | 1196/2562 [1:00:57<1:05:13,  2.86s/it]
 47%|████▋     | 1197/2562 [1:01:00<1:05:08,  2.86s/it]
 47%|████▋     | 1198/2562 [1:01:03<1:06:10,  2.91s/it]
 47%|████▋     | 1199/2562 [1:01:06<1:05:46,  2.90s/it]
 47%|████▋     | 1200/2562 [1:01:08<1:04:54,  2.86s/it]
                                                       
{'loss': 0.9225, 'grad_norm': 4.983480453491211, 'learning_rate': 1.186772460599523e-05, 'epoch': 1.4}

 47%|████▋     | 1200/2562 [1:01:08<1:04:54,  2.86s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.59it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.73it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.19it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.88it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.55it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.78it/s]

 11%|█         | 15/140 [00:01<00:12,  9.81it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.81it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.85it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.63it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.55it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.65it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.41it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.36it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.49it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.37it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.07it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.14it/s]

 20%|██        | 28/140 [00:02<00:11,  9.49it/s]

 21%|██        | 29/140 [00:02<00:11,  9.34it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.02it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.03it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.26it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.73it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.81it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.83it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.93it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.94it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 30%|███       | 42/140 [00:04<00:09,  9.90it/s]

 31%|███       | 43/140 [00:04<00:09,  9.88it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.85it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.85it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.57it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.19it/s]

 34%|███▍      | 48/140 [00:04<00:10,  9.18it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.10it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.21it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.40it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.43it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.45it/s]

 40%|████      | 56/140 [00:05<00:08,  9.69it/s]

 41%|████      | 57/140 [00:05<00:08,  9.49it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.39it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.48it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.59it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.81it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.62it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.56it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.54it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.51it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.73it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.79it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.81it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.80it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.86it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.85it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.81it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.59it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.22it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.34it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.28it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.19it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.09it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.10it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.31it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.19it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.17it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.07it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.09it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.99it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.06it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.26it/s]

 66%|██████▋   | 93/140 [00:09<00:05,  9.40it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.25it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.00it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.09it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.34it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.22it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.31it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.48it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.30it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.37it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.55it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.76it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.62it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.51it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.58it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.65it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.83it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.85it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.52it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.53it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.27it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.40it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.11it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.06it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.21it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.35it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.24it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.03it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.82it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.57it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.68it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  8.99it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.48it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.60it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.64it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.45it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.48it/s]

100%|██████████| 140/140 [00:14<00:00,  8.09it/s]
                                                       

                                                 
{'eval_loss': 0.9971206784248352, 'eval_runtime': 15.1441, 'eval_samples_per_second': 18.489, 'eval_steps_per_second': 9.245, 'epoch': 1.4}

 47%|████▋     | 1200/2562 [1:01:24<1:04:54,  2.86s/it]

100%|██████████| 140/140 [00:14<00:00,  8.09it/s]

                                                 
 47%|████▋     | 1201/2562 [1:01:27<2:49:33,  7.48s/it]
 47%|████▋     | 1202/2562 [1:01:29<2:17:50,  6.08s/it]
 47%|████▋     | 1203/2562 [1:01:32<1:54:50,  5.07s/it]
 47%|████▋     | 1204/2562 [1:01:35<1:41:32,  4.49s/it]
 47%|████▋     | 1205/2562 [1:01:38<1:29:22,  3.95s/it]
 47%|████▋     | 1206/2562 [1:01:41<1:21:15,  3.60s/it]
 47%|████▋     | 1207/2562 [1:01:44<1:18:24,  3.47s/it]
 47%|████▋     | 1208/2562 [1:01:47<1:13:49,  3.27s/it]
 47%|████▋     | 1209/2562 [1:01:50<1:11:03,  3.15s/it]
 47%|████▋     | 1210/2562 [1:01:53<1:09:20,  3.08s/it]
                                                       
{'loss': 0.9507, 'grad_norm': 4.884205341339111, 'learning_rate': 1.1740720367303958e-05, 'epoch': 1.42}

 47%|████▋     | 1210/2562 [1:01:53<1:09:20,  3.08s/it]
 47%|████▋     | 1211/2562 [1:01:55<1:07:14,  2.99s/it]
 47%|████▋     | 1212/2562 [1:01:58<1:05:59,  2.93s/it]
 47%|████▋     | 1213/2562 [1:02:01<1:06:11,  2.94s/it]
 47%|████▋     | 1214/2562 [1:02:04<1:05:23,  2.91s/it]
 47%|████▋     | 1215/2562 [1:02:07<1:05:40,  2.93s/it]
 47%|████▋     | 1216/2562 [1:02:10<1:05:51,  2.94s/it]
 48%|████▊     | 1217/2562 [1:02:13<1:05:14,  2.91s/it]
 48%|████▊     | 1218/2562 [1:02:16<1:04:50,  2.89s/it]
 48%|████▊     | 1219/2562 [1:02:19<1:05:46,  2.94s/it]
 48%|████▊     | 1220/2562 [1:02:21<1:05:25,  2.93s/it]
                                                       
{'loss': 0.9197, 'grad_norm': 4.969295501708984, 'learning_rate': 1.1613425901150595e-05, 'epoch': 1.43}

 48%|████▊     | 1220/2562 [1:02:21<1:05:25,  2.93s/it]
 48%|████▊     | 1221/2562 [1:02:24<1:05:27,  2.93s/it]
 48%|████▊     | 1222/2562 [1:02:27<1:06:09,  2.96s/it]
 48%|████▊     | 1223/2562 [1:02:30<1:05:55,  2.95s/it]
 48%|████▊     | 1224/2562 [1:02:33<1:05:12,  2.92s/it]
 48%|████▊     | 1225/2562 [1:02:36<1:05:52,  2.96s/it]
 48%|████▊     | 1226/2562 [1:02:39<1:05:05,  2.92s/it]
 48%|████▊     | 1227/2562 [1:02:42<1:04:41,  2.91s/it]
 48%|████▊     | 1228/2562 [1:02:45<1:04:43,  2.91s/it]
 48%|████▊     | 1229/2562 [1:02:48<1:04:26,  2.90s/it]
 48%|████▊     | 1230/2562 [1:02:51<1:04:21,  2.90s/it]
                                                       
{'loss': 0.9441, 'grad_norm': 4.597759246826172, 'learning_rate': 1.1485862431133445e-05, 'epoch': 1.44}

 48%|████▊     | 1230/2562 [1:02:51<1:04:21,  2.90s/it]
 48%|████▊     | 1231/2562 [1:02:54<1:04:24,  2.90s/it]
 48%|████▊     | 1232/2562 [1:02:56<1:04:02,  2.89s/it]
 48%|████▊     | 1233/2562 [1:02:59<1:03:44,  2.88s/it]
 48%|████▊     | 1234/2562 [1:03:02<1:05:02,  2.94s/it]
 48%|████▊     | 1235/2562 [1:03:05<1:04:26,  2.91s/it]
 48%|████▊     | 1236/2562 [1:03:08<1:04:28,  2.92s/it]
 48%|████▊     | 1237/2562 [1:03:11<1:05:28,  2.97s/it]
 48%|████▊     | 1238/2562 [1:03:14<1:04:41,  2.93s/it]
 48%|████▊     | 1239/2562 [1:03:17<1:04:54,  2.94s/it]
 48%|████▊     | 1240/2562 [1:03:20<1:03:18,  2.87s/it]
                                                       
{'loss': 0.9052, 'grad_norm': 4.90260648727417, 'learning_rate': 1.1358051225701404e-05, 'epoch': 1.45}

 48%|████▊     | 1240/2562 [1:03:20<1:03:18,  2.87s/it]
 48%|████▊     | 1241/2562 [1:03:23<1:02:31,  2.84s/it]
 48%|████▊     | 1242/2562 [1:03:26<1:04:20,  2.92s/it]
 49%|████▊     | 1243/2562 [1:03:28<1:03:39,  2.90s/it]
 49%|████▊     | 1244/2562 [1:03:31<1:03:33,  2.89s/it]
 49%|████▊     | 1245/2562 [1:03:34<1:03:40,  2.90s/it]
 49%|████▊     | 1246/2562 [1:03:37<1:03:11,  2.88s/it]
 49%|████▊     | 1247/2562 [1:03:40<1:02:55,  2.87s/it]
 49%|████▊     | 1248/2562 [1:03:43<1:03:08,  2.88s/it]
 49%|████▉     | 1249/2562 [1:03:46<1:02:41,  2.86s/it]
 49%|████▉     | 1250/2562 [1:03:49<1:02:59,  2.88s/it]
                                                       
{'loss': 0.9167, 'grad_norm': 4.716863632202148, 'learning_rate': 1.1230013594607874e-05, 'epoch': 1.46}

 49%|████▉     | 1250/2562 [1:03:49<1:02:59,  2.88s/it]
 49%|████▉     | 1251/2562 [1:03:52<1:04:42,  2.96s/it]
 49%|████▉     | 1252/2562 [1:03:55<1:04:16,  2.94s/it]
 49%|████▉     | 1253/2562 [1:03:58<1:03:45,  2.92s/it]
 49%|████▉     | 1254/2562 [1:04:01<1:04:23,  2.95s/it]
 49%|████▉     | 1255/2562 [1:04:03<1:03:11,  2.90s/it]
 49%|████▉     | 1256/2562 [1:04:06<1:03:36,  2.92s/it]
 49%|████▉     | 1257/2562 [1:04:09<1:02:31,  2.87s/it]
 49%|████▉     | 1258/2562 [1:04:12<1:01:32,  2.83s/it]
 49%|████▉     | 1259/2562 [1:04:15<1:02:02,  2.86s/it]
 49%|████▉     | 1260/2562 [1:04:18<1:02:23,  2.88s/it]
                                                       
{'loss': 0.9307, 'grad_norm': 4.635587692260742, 'learning_rate': 1.1101770885357843e-05, 'epoch': 1.47}

 49%|████▉     | 1260/2562 [1:04:18<1:02:23,  2.88s/it]
 49%|████▉     | 1261/2562 [1:04:21<1:02:51,  2.90s/it]
 49%|████▉     | 1262/2562 [1:04:24<1:03:37,  2.94s/it]
 49%|████▉     | 1263/2562 [1:04:26<1:02:59,  2.91s/it]
 49%|████▉     | 1264/2562 [1:04:29<1:03:18,  2.93s/it]
 49%|████▉     | 1265/2562 [1:04:32<1:03:37,  2.94s/it]
 49%|████▉     | 1266/2562 [1:04:35<1:02:36,  2.90s/it]
 49%|████▉     | 1267/2562 [1:04:38<1:02:06,  2.88s/it]
 49%|████▉     | 1268/2562 [1:04:41<1:03:00,  2.92s/it]
 50%|████▉     | 1269/2562 [1:04:44<1:02:37,  2.91s/it]
 50%|████▉     | 1270/2562 [1:04:47<1:01:33,  2.86s/it]
                                                       
{'loss': 0.8939, 'grad_norm': 4.952556133270264, 'learning_rate': 1.0973344479648652e-05, 'epoch': 1.49}

 50%|████▉     | 1270/2562 [1:04:47<1:01:33,  2.86s/it]
 50%|████▉     | 1271/2562 [1:04:50<1:01:34,  2.86s/it]
 50%|████▉     | 1272/2562 [1:04:52<1:01:37,  2.87s/it]
 50%|████▉     | 1273/2562 [1:04:55<1:01:47,  2.88s/it]
 50%|████▉     | 1274/2562 [1:04:58<1:01:59,  2.89s/it]
 50%|████▉     | 1275/2562 [1:05:01<1:00:49,  2.84s/it]
 50%|████▉     | 1276/2562 [1:05:04<1:00:18,  2.81s/it]
 50%|████▉     | 1277/2562 [1:05:07<1:01:48,  2.89s/it]
 50%|████▉     | 1278/2562 [1:05:10<1:02:06,  2.90s/it]
 50%|████▉     | 1279/2562 [1:05:13<1:01:57,  2.90s/it]
 50%|████▉     | 1280/2562 [1:05:16<1:02:02,  2.90s/it]
                                                       
{'loss': 0.9088, 'grad_norm': 5.0410075187683105, 'learning_rate': 1.0844755789805042e-05, 'epoch': 1.5}

 50%|████▉     | 1280/2562 [1:05:16<1:02:02,  2.90s/it]
 50%|█████     | 1281/2562 [1:05:18<1:01:01,  2.86s/it]
 50%|█████     | 1282/2562 [1:05:21<1:02:01,  2.91s/it]
 50%|█████     | 1283/2562 [1:05:24<1:03:17,  2.97s/it]
 50%|█████     | 1284/2562 [1:05:27<1:02:46,  2.95s/it]
 50%|█████     | 1285/2562 [1:05:30<1:02:46,  2.95s/it]
 50%|█████     | 1286/2562 [1:05:33<1:02:55,  2.96s/it]
 50%|█████     | 1287/2562 [1:05:36<1:01:59,  2.92s/it]
 50%|█████     | 1288/2562 [1:05:39<1:01:03,  2.88s/it]
 50%|█████     | 1289/2562 [1:05:42<1:01:53,  2.92s/it]
 50%|█████     | 1290/2562 [1:05:45<1:03:15,  2.98s/it]
                                                       
{'loss': 0.9048, 'grad_norm': 4.540219306945801, 'learning_rate': 1.0716026255209124e-05, 'epoch': 1.51}

 50%|█████     | 1290/2562 [1:05:45<1:03:15,  2.98s/it]
 50%|█████     | 1291/2562 [1:05:48<1:01:38,  2.91s/it]
 50%|█████     | 1292/2562 [1:05:51<1:02:16,  2.94s/it]
 50%|█████     | 1293/2562 [1:05:54<1:01:50,  2.92s/it]
 51%|█████     | 1294/2562 [1:05:56<1:00:52,  2.88s/it]
 51%|█████     | 1295/2562 [1:05:59<1:00:44,  2.88s/it]
 51%|█████     | 1296/2562 [1:06:02<1:01:14,  2.90s/it]
 51%|█████     | 1297/2562 [1:06:05<1:01:07,  2.90s/it]
 51%|█████     | 1298/2562 [1:06:08<1:00:55,  2.89s/it]
 51%|█████     | 1299/2562 [1:06:11<1:01:20,  2.91s/it]
 51%|█████     | 1300/2562 [1:06:14<1:01:49,  2.94s/it]
                                                       
{'loss': 0.8954, 'grad_norm': 4.905197620391846, 'learning_rate': 1.0587177338725834e-05, 'epoch': 1.52}

 51%|█████     | 1300/2562 [1:06:14<1:01:49,  2.94s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.83it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.78it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.22it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.89it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.55it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

 10%|█         | 14/140 [00:01<00:12,  9.75it/s]

 11%|█         | 15/140 [00:01<00:12,  9.77it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.76it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.78it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.57it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.52it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.46it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.42it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.54it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.43it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.14it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.19it/s]

 20%|██        | 28/140 [00:02<00:11,  9.51it/s]

 21%|██        | 29/140 [00:02<00:11,  9.36it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.06it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.06it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.25it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.70it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.79it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.83it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.84it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.91it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 31%|███       | 43/140 [00:04<00:09,  9.99it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.96it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.95it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.69it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.35it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.28it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.19it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.25it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.45it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.45it/s]

 40%|████      | 56/140 [00:05<00:08,  9.72it/s]

 41%|████      | 57/140 [00:05<00:08,  9.52it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.42it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.49it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.59it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.82it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.62it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.57it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.54it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.51it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.73it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.81it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.88it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.88it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.86it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.66it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.29it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.40it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.33it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.25it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.15it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.15it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.34it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.24it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.20it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.10it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.13it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.03it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.10it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.28it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.45it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.29it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.04it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.13it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.29it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.35it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.49it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.32it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.38it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.71it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.73it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.61it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.51it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.57it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.64it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.68it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.84it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.84it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.53it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.54it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.27it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.40it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.11it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.08it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.24it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.41it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.31it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.08it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.82it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.59it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.73it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.06it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.30it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.66it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.70it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.51it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.53it/s]

100%|██████████| 140/140 [00:14<00:00,  8.08it/s]
                                                       

                                                 
{'eval_loss': 0.9798517227172852, 'eval_runtime': 15.1149, 'eval_samples_per_second': 18.525, 'eval_steps_per_second': 9.262, 'epoch': 1.52}

 51%|█████     | 1300/2562 [1:06:29<1:01:49,  2.94s/it]

100%|██████████| 140/140 [00:14<00:00,  8.08it/s]

                                                 
 51%|█████     | 1301/2562 [1:06:32<2:37:15,  7.48s/it]
 51%|█████     | 1302/2562 [1:06:35<2:08:04,  6.10s/it]
 51%|█████     | 1303/2562 [1:06:38<1:48:29,  5.17s/it]
 51%|█████     | 1304/2562 [1:06:41<1:33:51,  4.48s/it]
 51%|█████     | 1305/2562 [1:06:44<1:23:37,  3.99s/it]
 51%|█████     | 1306/2562 [1:06:47<1:16:57,  3.68s/it]
 51%|█████     | 1307/2562 [1:06:50<1:12:36,  3.47s/it]
 51%|█████     | 1308/2562 [1:06:52<1:08:54,  3.30s/it]
 51%|█████     | 1309/2562 [1:06:56<1:07:26,  3.23s/it]
 51%|█████     | 1310/2562 [1:06:58<1:05:12,  3.13s/it]
                                                       
{'loss': 0.8879, 'grad_norm': 4.816926956176758, 'learning_rate': 1.0458230523124443e-05, 'epoch': 1.53}

 51%|█████     | 1310/2562 [1:06:58<1:05:12,  3.13s/it]
 51%|█████     | 1311/2562 [1:07:01<1:03:34,  3.05s/it]
 51%|█████     | 1312/2562 [1:07:04<1:03:54,  3.07s/it]
 51%|█████     | 1313/2562 [1:07:07<1:02:15,  2.99s/it]
 51%|█████▏    | 1314/2562 [1:07:10<1:00:19,  2.90s/it]
 51%|█████▏    | 1315/2562 [1:07:13<1:01:01,  2.94s/it]
 51%|█████▏    | 1316/2562 [1:07:16<1:00:59,  2.94s/it]
 51%|█████▏    | 1317/2562 [1:07:19<1:00:15,  2.90s/it]
 51%|█████▏    | 1318/2562 [1:07:22<1:01:08,  2.95s/it]
 51%|█████▏    | 1319/2562 [1:07:25<1:00:18,  2.91s/it]
 52%|█████▏    | 1320/2562 [1:07:27<59:17,  2.86s/it]  
                                                     
{'loss': 0.9483, 'grad_norm': 5.098050117492676, 'learning_rate': 1.0329207307496785e-05, 'epoch': 1.55}

 52%|█████▏    | 1320/2562 [1:07:27<59:17,  2.86s/it]
 52%|█████▏    | 1321/2562 [1:07:30<1:00:47,  2.94s/it]
 52%|█████▏    | 1322/2562 [1:07:33<59:21,  2.87s/it]  
 52%|█████▏    | 1323/2562 [1:07:36<1:00:06,  2.91s/it]
 52%|█████▏    | 1324/2562 [1:07:39<1:00:14,  2.92s/it]
 52%|█████▏    | 1325/2562 [1:07:42<59:06,  2.87s/it]  
 52%|█████▏    | 1326/2562 [1:07:45<58:43,  2.85s/it]
 52%|█████▏    | 1327/2562 [1:07:48<58:52,  2.86s/it]
 52%|█████▏    | 1328/2562 [1:07:50<58:32,  2.85s/it]
 52%|█████▏    | 1329/2562 [1:07:53<57:58,  2.82s/it]
 52%|█████▏    | 1330/2562 [1:07:56<58:57,  2.87s/it]
                                                     
{'loss': 0.9579, 'grad_norm': 5.023376941680908, 'learning_rate': 1.0200129203672754e-05, 'epoch': 1.56}

 52%|█████▏    | 1330/2562 [1:07:56<58:57,  2.87s/it]
 52%|█████▏    | 1331/2562 [1:07:59<58:25,  2.85s/it]
 52%|█████▏    | 1332/2562 [1:08:02<58:39,  2.86s/it]
 52%|█████▏    | 1333/2562 [1:08:05<59:13,  2.89s/it]
 52%|█████▏    | 1334/2562 [1:08:07<58:23,  2.85s/it]
 52%|█████▏    | 1335/2562 [1:08:10<58:18,  2.85s/it]
 52%|█████▏    | 1336/2562 [1:08:13<59:14,  2.90s/it]
 52%|█████▏    | 1337/2562 [1:08:16<58:52,  2.88s/it]
 52%|█████▏    | 1338/2562 [1:08:19<58:38,  2.87s/it]
 52%|█████▏    | 1339/2562 [1:08:22<59:15,  2.91s/it]
 52%|█████▏    | 1340/2562 [1:08:25<58:22,  2.87s/it]
                                                     
{'loss': 0.9202, 'grad_norm': 5.138041973114014, 'learning_rate': 1.007101773263365e-05, 'epoch': 1.57}

 52%|█████▏    | 1340/2562 [1:08:25<58:22,  2.87s/it]
 52%|█████▏    | 1341/2562 [1:08:28<58:52,  2.89s/it]
 52%|█████▏    | 1342/2562 [1:08:31<58:43,  2.89s/it]
 52%|█████▏    | 1343/2562 [1:08:33<58:10,  2.86s/it]
 52%|█████▏    | 1344/2562 [1:08:36<58:36,  2.89s/it]
 52%|█████▏    | 1345/2562 [1:08:39<59:28,  2.93s/it]
 53%|█████▎    | 1346/2562 [1:08:42<59:39,  2.94s/it]
 53%|█████▎    | 1347/2562 [1:08:45<58:59,  2.91s/it]
 53%|█████▎    | 1348/2562 [1:08:48<1:00:00,  2.97s/it]
 53%|█████▎    | 1349/2562 [1:08:51<1:00:10,  2.98s/it]
 53%|█████▎    | 1350/2562 [1:08:54<1:00:48,  3.01s/it]
                                                       
{'loss': 0.9401, 'grad_norm': 5.197987079620361, 'learning_rate': 9.941894420924044e-06, 'epoch': 1.58}

 53%|█████▎    | 1350/2562 [1:08:54<1:00:48,  3.01s/it]
 53%|█████▎    | 1351/2562 [1:08:57<59:27,  2.95s/it]  
 53%|█████▎    | 1352/2562 [1:09:00<58:37,  2.91s/it]
 53%|█████▎    | 1353/2562 [1:09:03<58:35,  2.91s/it]
 53%|█████▎    | 1354/2562 [1:09:06<58:11,  2.89s/it]
 53%|█████▎    | 1355/2562 [1:09:09<58:10,  2.89s/it]
 53%|█████▎    | 1356/2562 [1:09:12<58:45,  2.92s/it]
 53%|█████▎    | 1357/2562 [1:09:15<58:07,  2.89s/it]
 53%|█████▎    | 1358/2562 [1:09:17<57:40,  2.87s/it]
 53%|█████▎    | 1359/2562 [1:09:21<59:34,  2.97s/it]
 53%|█████▎    | 1360/2562 [1:09:23<58:51,  2.94s/it]
                                                     
{'loss': 0.8931, 'grad_norm': 4.839508056640625, 'learning_rate': 9.812780797062678e-06, 'epoch': 1.59}

 53%|█████▎    | 1360/2562 [1:09:23<58:51,  2.94s/it]
 53%|█████▎    | 1361/2562 [1:09:26<58:38,  2.93s/it]
 53%|█████▎    | 1362/2562 [1:09:29<58:54,  2.95s/it]
 53%|█████▎    | 1363/2562 [1:09:32<59:14,  2.96s/it]
 53%|█████▎    | 1364/2562 [1:09:35<58:38,  2.94s/it]
 53%|█████▎    | 1365/2562 [1:09:38<59:01,  2.96s/it]
 53%|█████▎    | 1366/2562 [1:09:41<59:25,  2.98s/it]
 53%|█████▎    | 1367/2562 [1:09:44<58:39,  2.94s/it]
 53%|█████▎    | 1368/2562 [1:09:47<58:51,  2.96s/it]
 53%|█████▎    | 1369/2562 [1:09:50<58:14,  2.93s/it]
 53%|█████▎    | 1370/2562 [1:09:53<58:38,  2.95s/it]
                                                     
{'loss': 0.9051, 'grad_norm': 4.684476375579834, 'learning_rate': 9.68369838795306e-06, 'epoch': 1.6}

 53%|█████▎    | 1370/2562 [1:09:53<58:38,  2.95s/it]
 54%|█████▎    | 1371/2562 [1:09:56<59:08,  2.98s/it]
 54%|█████▎    | 1372/2562 [1:09:59<58:11,  2.93s/it]
 54%|█████▎    | 1373/2562 [1:10:02<57:15,  2.89s/it]
 54%|█████▎    | 1374/2562 [1:10:05<58:26,  2.95s/it]
 54%|█████▎    | 1375/2562 [1:10:07<57:10,  2.89s/it]
 54%|█████▎    | 1376/2562 [1:10:10<57:09,  2.89s/it]
 54%|█████▎    | 1377/2562 [1:10:13<57:37,  2.92s/it]
 54%|█████▍    | 1378/2562 [1:10:16<57:10,  2.90s/it]
 54%|█████▍    | 1379/2562 [1:10:19<57:08,  2.90s/it]
 54%|█████▍    | 1380/2562 [1:10:22<57:13,  2.91s/it]
                                                     
{'loss': 0.9124, 'grad_norm': 4.945702075958252, 'learning_rate': 9.554668715294305e-06, 'epoch': 1.62}

 54%|█████▍    | 1380/2562 [1:10:22<57:13,  2.91s/it]
 54%|█████▍    | 1381/2562 [1:10:25<57:01,  2.90s/it]
 54%|█████▍    | 1382/2562 [1:10:28<56:10,  2.86s/it]
 54%|█████▍    | 1383/2562 [1:10:31<56:45,  2.89s/it]
 54%|█████▍    | 1384/2562 [1:10:33<55:48,  2.84s/it]
 54%|█████▍    | 1385/2562 [1:10:36<55:12,  2.81s/it]
 54%|█████▍    | 1386/2562 [1:10:39<56:12,  2.87s/it]
 54%|█████▍    | 1387/2562 [1:10:42<56:13,  2.87s/it]
 54%|█████▍    | 1388/2562 [1:10:45<57:11,  2.92s/it]
 54%|█████▍    | 1389/2562 [1:10:48<57:01,  2.92s/it]
 54%|█████▍    | 1390/2562 [1:10:51<57:12,  2.93s/it]
                                                     
{'loss': 0.9096, 'grad_norm': 4.585940837860107, 'learning_rate': 9.425713291992878e-06, 'epoch': 1.63}

 54%|█████▍    | 1390/2562 [1:10:51<57:12,  2.93s/it]
 54%|█████▍    | 1391/2562 [1:10:54<56:43,  2.91s/it]
 54%|█████▍    | 1392/2562 [1:10:57<56:48,  2.91s/it]
 54%|█████▍    | 1393/2562 [1:10:59<56:12,  2.89s/it]
 54%|█████▍    | 1394/2562 [1:11:03<57:24,  2.95s/it]
 54%|█████▍    | 1395/2562 [1:11:05<57:11,  2.94s/it]
 54%|█████▍    | 1396/2562 [1:11:08<56:21,  2.90s/it]
 55%|█████▍    | 1397/2562 [1:11:11<58:08,  2.99s/it]
 55%|█████▍    | 1398/2562 [1:11:14<57:02,  2.94s/it]
 55%|█████▍    | 1399/2562 [1:11:17<56:58,  2.94s/it]
 55%|█████▍    | 1400/2562 [1:11:20<57:09,  2.95s/it]
                                                     
{'loss': 0.8814, 'grad_norm': 4.938027381896973, 'learning_rate': 9.296853618575753e-06, 'epoch': 1.64}

 55%|█████▍    | 1400/2562 [1:11:20<57:09,  2.95s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 19.07it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.76it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.21it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.88it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.56it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.75it/s]

 10%|█         | 14/140 [00:01<00:12,  9.79it/s]

 11%|█         | 15/140 [00:01<00:12,  9.81it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.81it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.83it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.62it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.51it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.63it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.38it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.33it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.46it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.31it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.02it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.13it/s]

 20%|██        | 28/140 [00:02<00:11,  9.49it/s]

 21%|██        | 29/140 [00:02<00:11,  9.35it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.04it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.05it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.25it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.72it/s]

 25%|██▌       | 35/140 [00:03<00:10,  9.78it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.80it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.83it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.87it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.94it/s]

 29%|██▉       | 41/140 [00:04<00:10,  9.90it/s]

 30%|███       | 42/140 [00:04<00:09,  9.91it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.92it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.87it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.64it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.31it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.28it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.18it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.24it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.45it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.46it/s]

 40%|████      | 56/140 [00:05<00:08,  9.72it/s]

 41%|████      | 57/140 [00:05<00:08,  9.49it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.38it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.48it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.58it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.79it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.62it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.55it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.54it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.52it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.74it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.79it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.81it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.83it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.83it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.83it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.84it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.59it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.17it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.34it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.31it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.19it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.12it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.14it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.33it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.21it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.18it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.08it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.99it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.06it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.25it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.42it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.27it/s]

 68%|██████▊   | 95/140 [00:09<00:05,  9.00it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.11it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.29it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.34it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.50it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.32it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.39it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.55it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.76it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.62it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.51it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.58it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.65it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.82it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.82it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.51it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.39it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.52it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▌ | 120/140 [00:12<00:02,  9.47it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.35it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.02it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.02it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.22it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.19it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.36it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.22it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.01it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.82it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.58it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.69it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  9.02it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.26it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.61it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.64it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.45it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.49it/s]

100%|██████████| 140/140 [00:14<00:00,  7.96it/s]
                                                     

                                                 
{'eval_loss': 0.9684665203094482, 'eval_runtime': 15.1514, 'eval_samples_per_second': 18.48, 'eval_steps_per_second': 9.24, 'epoch': 1.64}

 55%|█████▍    | 1400/2562 [1:11:35<57:09,  2.95s/it]

100%|██████████| 140/140 [00:14<00:00,  7.96it/s]

                                                 
 55%|█████▍    | 1401/2562 [1:11:38<2:25:06,  7.50s/it]
 55%|█████▍    | 1402/2562 [1:11:41<1:58:12,  6.11s/it]
 55%|█████▍    | 1403/2562 [1:11:44<1:39:48,  5.17s/it]
 55%|█████▍    | 1404/2562 [1:11:47<1:27:03,  4.51s/it]
 55%|█████▍    | 1405/2562 [1:11:50<1:17:38,  4.03s/it]
 55%|█████▍    | 1406/2562 [1:11:53<1:10:56,  3.68s/it]
 55%|█████▍    | 1407/2562 [1:11:56<1:05:41,  3.41s/it]
 55%|█████▍    | 1408/2562 [1:11:59<1:02:45,  3.26s/it]
 55%|█████▍    | 1409/2562 [1:12:01<1:00:27,  3.15s/it]
 55%|█████▌    | 1410/2562 [1:12:04<58:28,  3.05s/it]  
                                                     
{'loss': 0.9149, 'grad_norm': 5.1518378257751465, 'learning_rate': 9.1681111796057e-06, 'epoch': 1.65}

 55%|█████▌    | 1410/2562 [1:12:04<58:28,  3.05s/it]
 55%|█████▌    | 1411/2562 [1:12:07<56:42,  2.96s/it]
 55%|█████▌    | 1412/2562 [1:12:10<56:37,  2.95s/it]
 55%|█████▌    | 1413/2562 [1:12:13<56:42,  2.96s/it]
 55%|█████▌    | 1414/2562 [1:12:16<55:34,  2.90s/it]
 55%|█████▌    | 1415/2562 [1:12:19<56:22,  2.95s/it]
 55%|█████▌    | 1416/2562 [1:12:22<55:58,  2.93s/it]
 55%|█████▌    | 1417/2562 [1:12:25<55:31,  2.91s/it]
 55%|█████▌    | 1418/2562 [1:12:28<56:21,  2.96s/it]
 55%|█████▌    | 1419/2562 [1:12:31<56:12,  2.95s/it]
 55%|█████▌    | 1420/2562 [1:12:33<55:27,  2.91s/it]
                                                     
{'loss': 0.9313, 'grad_norm': 5.210409641265869, 'learning_rate': 9.039507440099164e-06, 'epoch': 1.66}

 55%|█████▌    | 1420/2562 [1:12:33<55:27,  2.91s/it]
 55%|█████▌    | 1421/2562 [1:12:36<55:56,  2.94s/it]
 56%|█████▌    | 1422/2562 [1:12:39<55:10,  2.90s/it]
 56%|█████▌    | 1423/2562 [1:12:42<54:48,  2.89s/it]
 56%|█████▌    | 1424/2562 [1:12:45<54:51,  2.89s/it]
 56%|█████▌    | 1425/2562 [1:12:48<54:21,  2.87s/it]
 56%|█████▌    | 1426/2562 [1:12:51<54:11,  2.86s/it]
 56%|█████▌    | 1427/2562 [1:12:54<54:28,  2.88s/it]
 56%|█████▌    | 1428/2562 [1:12:56<53:30,  2.83s/it]
 56%|█████▌    | 1429/2562 [1:12:59<53:37,  2.84s/it]
 56%|█████▌    | 1430/2562 [1:13:02<53:48,  2.85s/it]
                                                     
{'loss': 0.8852, 'grad_norm': 4.617379188537598, 'learning_rate': 8.911063841947476e-06, 'epoch': 1.67}

 56%|█████▌    | 1430/2562 [1:13:02<53:48,  2.85s/it]
 56%|█████▌    | 1431/2562 [1:13:05<53:21,  2.83s/it]
 56%|█████▌    | 1432/2562 [1:13:08<54:28,  2.89s/it]
 56%|█████▌    | 1433/2562 [1:13:11<53:55,  2.87s/it]
 56%|█████▌    | 1434/2562 [1:13:13<53:34,  2.85s/it]
 56%|█████▌    | 1435/2562 [1:13:16<54:11,  2.89s/it]
 56%|█████▌    | 1436/2562 [1:13:19<54:06,  2.88s/it]
 56%|█████▌    | 1437/2562 [1:13:22<54:22,  2.90s/it]
 56%|█████▌    | 1438/2562 [1:13:25<54:49,  2.93s/it]
 56%|█████▌    | 1439/2562 [1:13:28<54:12,  2.90s/it]
 56%|█████▌    | 1440/2562 [1:13:31<54:11,  2.90s/it]
                                                     
{'loss': 0.8684, 'grad_norm': 5.197981357574463, 'learning_rate': 8.78280180034184e-06, 'epoch': 1.69}

 56%|█████▌    | 1440/2562 [1:13:31<54:11,  2.90s/it]
 56%|█████▌    | 1441/2562 [1:13:34<54:33,  2.92s/it]
 56%|█████▋    | 1442/2562 [1:13:37<54:12,  2.90s/it]
 56%|█████▋    | 1443/2562 [1:13:40<54:15,  2.91s/it]
 56%|█████▋    | 1444/2562 [1:13:43<54:58,  2.95s/it]
 56%|█████▋    | 1445/2562 [1:13:45<53:39,  2.88s/it]
 56%|█████▋    | 1446/2562 [1:13:48<52:52,  2.84s/it]
 56%|█████▋    | 1447/2562 [1:13:51<54:21,  2.92s/it]
 57%|█████▋    | 1448/2562 [1:13:54<53:38,  2.89s/it]
 57%|█████▋    | 1449/2562 [1:13:57<53:47,  2.90s/it]
 57%|█████▋    | 1450/2562 [1:14:00<54:39,  2.95s/it]
                                                     
{'loss': 0.8863, 'grad_norm': 5.082547187805176, 'learning_rate': 8.654742700202849e-06, 'epoch': 1.7}

 57%|█████▋    | 1450/2562 [1:14:00<54:39,  2.95s/it]
 57%|█████▋    | 1451/2562 [1:14:03<53:21,  2.88s/it]
 57%|█████▋    | 1452/2562 [1:14:06<53:03,  2.87s/it]
 57%|█████▋    | 1453/2562 [1:14:09<54:02,  2.92s/it]
 57%|█████▋    | 1454/2562 [1:14:11<52:45,  2.86s/it]
 57%|█████▋    | 1455/2562 [1:14:14<52:12,  2.83s/it]
 57%|█████▋    | 1456/2562 [1:14:17<52:52,  2.87s/it]
 57%|█████▋    | 1457/2562 [1:14:20<53:13,  2.89s/it]
 57%|█████▋    | 1458/2562 [1:14:23<53:11,  2.89s/it]
 57%|█████▋    | 1459/2562 [1:14:26<53:07,  2.89s/it]
 57%|█████▋    | 1460/2562 [1:14:29<52:38,  2.87s/it]
                                                     
{'loss': 0.8963, 'grad_norm': 5.275082588195801, 'learning_rate': 8.526907892614986e-06, 'epoch': 1.71}

 57%|█████▋    | 1460/2562 [1:14:29<52:38,  2.87s/it]
 57%|█████▋    | 1461/2562 [1:14:31<52:03,  2.84s/it]
 57%|█████▋    | 1462/2562 [1:14:35<53:25,  2.91s/it]
 57%|█████▋    | 1463/2562 [1:14:37<53:05,  2.90s/it]
 57%|█████▋    | 1464/2562 [1:14:40<52:21,  2.86s/it]
 57%|█████▋    | 1465/2562 [1:14:43<53:21,  2.92s/it]
 57%|█████▋    | 1466/2562 [1:14:46<52:04,  2.85s/it]
 57%|█████▋    | 1467/2562 [1:14:49<51:40,  2.83s/it]
 57%|█████▋    | 1468/2562 [1:14:52<51:49,  2.84s/it]
 57%|█████▋    | 1469/2562 [1:14:54<51:35,  2.83s/it]
 57%|█████▋    | 1470/2562 [1:14:57<50:55,  2.80s/it]
                                                     
{'loss': 0.8778, 'grad_norm': 4.768045425415039, 'learning_rate': 8.399318691266806e-06, 'epoch': 1.72}

 57%|█████▋    | 1470/2562 [1:14:57<50:55,  2.80s/it]
 57%|█████▋    | 1471/2562 [1:15:00<53:09,  2.92s/it]
 57%|█████▋    | 1472/2562 [1:15:03<53:01,  2.92s/it]
 57%|█████▋    | 1473/2562 [1:15:06<52:51,  2.91s/it]
 58%|█████▊    | 1474/2562 [1:15:09<53:47,  2.97s/it]
 58%|█████▊    | 1475/2562 [1:15:12<53:13,  2.94s/it]
 58%|█████▊    | 1476/2562 [1:15:15<53:36,  2.96s/it]
 58%|█████▊    | 1477/2562 [1:15:18<52:30,  2.90s/it]
 58%|█████▊    | 1478/2562 [1:15:21<53:01,  2.93s/it]
 58%|█████▊    | 1479/2562 [1:15:24<53:31,  2.97s/it]
 58%|█████▊    | 1480/2562 [1:15:27<53:23,  2.96s/it]
                                                     
{'loss': 0.9029, 'grad_norm': 4.882233142852783, 'learning_rate': 8.271996368897345e-06, 'epoch': 1.73}

 58%|█████▊    | 1480/2562 [1:15:27<53:23,  2.96s/it]
 58%|█████▊    | 1481/2562 [1:15:30<52:49,  2.93s/it]
 58%|█████▊    | 1482/2562 [1:15:33<53:06,  2.95s/it]
 58%|█████▊    | 1483/2562 [1:15:35<52:02,  2.89s/it]
 58%|█████▊    | 1484/2562 [1:15:38<52:07,  2.90s/it]
 58%|█████▊    | 1485/2562 [1:15:41<52:36,  2.93s/it]
 58%|█████▊    | 1486/2562 [1:15:44<51:57,  2.90s/it]
 58%|█████▊    | 1487/2562 [1:15:47<52:02,  2.90s/it]
 58%|█████▊    | 1488/2562 [1:15:50<53:11,  2.97s/it]
 58%|█████▊    | 1489/2562 [1:15:53<53:03,  2.97s/it]
 58%|█████▊    | 1490/2562 [1:15:56<51:52,  2.90s/it]
                                                     
{'loss': 0.881, 'grad_norm': 5.028517723083496, 'learning_rate': 8.144962153749331e-06, 'epoch': 1.74}

 58%|█████▊    | 1490/2562 [1:15:56<51:52,  2.90s/it]
 58%|█████▊    | 1491/2562 [1:15:59<52:57,  2.97s/it]
 58%|█████▊    | 1492/2562 [1:16:02<52:05,  2.92s/it]
 58%|█████▊    | 1493/2562 [1:16:05<51:58,  2.92s/it]
 58%|█████▊    | 1494/2562 [1:16:08<52:51,  2.97s/it]
 58%|█████▊    | 1495/2562 [1:16:11<52:04,  2.93s/it]
 58%|█████▊    | 1496/2562 [1:16:14<51:16,  2.89s/it]
 58%|█████▊    | 1497/2562 [1:16:16<51:10,  2.88s/it]
 58%|█████▊    | 1498/2562 [1:16:19<51:10,  2.89s/it]
 59%|█████▊    | 1499/2562 [1:16:22<51:09,  2.89s/it]
 59%|█████▊    | 1500/2562 [1:16:25<51:43,  2.92s/it]
                                                     
{'loss': 0.8852, 'grad_norm': 5.014894485473633, 'learning_rate': 8.01823722602986e-06, 'epoch': 1.76}

 59%|█████▊    | 1500/2562 [1:16:25<51:43,  2.92s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.96it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.75it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.19it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.86it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.55it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.78it/s]

 11%|█         | 15/140 [00:01<00:12,  9.80it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.78it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.81it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.61it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.53it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.65it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.43it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.38it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.51it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.36it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.08it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.16it/s]

 20%|██        | 28/140 [00:02<00:11,  9.49it/s]

 21%|██        | 29/140 [00:02<00:11,  9.37it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.06it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.07it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.27it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.74it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.83it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.85it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.93it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.92it/s]

 29%|██▉       | 41/140 [00:04<00:10,  9.90it/s]

 30%|███       | 42/140 [00:04<00:09,  9.90it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.92it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.90it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.65it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.33it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.29it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.21it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.28it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.45it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.46it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.47it/s]

 40%|████      | 56/140 [00:05<00:08,  9.73it/s]

 41%|████      | 57/140 [00:05<00:08,  9.51it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.42it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.60it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.82it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.65it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.57it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.56it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.52it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.72it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.86it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.88it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.85it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.83it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.57it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.21it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.36it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.31it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.22it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.13it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.14it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.35it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.22it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.20it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.09it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.11it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.00it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.08it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.27it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.42it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.26it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.03it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.13it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.37it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.26it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.31it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.46it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.27it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.32it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.68it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.72it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.59it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.49it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.57it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.62it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.65it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.85it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.86it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.56it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.43it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.54it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.34it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.31it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.43it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.13it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.11it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.28it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.24it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.39it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.27it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.04it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.82it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.57it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.70it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.01it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.27it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.64it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.68it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.52it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.54it/s]

100%|██████████| 140/140 [00:14<00:00,  8.13it/s]
                                                     

                                                 
{'eval_loss': 0.9540739059448242, 'eval_runtime': 15.1241, 'eval_samples_per_second': 18.514, 'eval_steps_per_second': 9.257, 'epoch': 1.76}

 59%|█████▊    | 1500/2562 [1:16:40<51:43,  2.92s/it]

100%|██████████| 140/140 [00:14<00:00,  8.13it/s]

                                                 /opt/conda/lib/python3.11/site-packages/peft/utils/save_and_load.py:372: UserWarning: Could not find a config file in  - will assume that the vocabulary was not modified.
  warnings.warn(

 59%|█████▊    | 1501/2562 [1:16:44<2:18:15,  7.82s/it]
 59%|█████▊    | 1502/2562 [1:16:47<1:52:17,  6.36s/it]
 59%|█████▊    | 1503/2562 [1:16:51<1:35:11,  5.39s/it]
 59%|█████▊    | 1504/2562 [1:16:53<1:21:36,  4.63s/it]
 59%|█████▊    | 1505/2562 [1:16:56<1:11:40,  4.07s/it]
 59%|█████▉    | 1506/2562 [1:16:59<1:06:02,  3.75s/it]
 59%|█████▉    | 1507/2562 [1:17:02<1:00:47,  3.46s/it]
 59%|█████▉    | 1508/2562 [1:17:05<57:44,  3.29s/it]  
 59%|█████▉    | 1509/2562 [1:17:08<56:14,  3.21s/it]
 59%|█████▉    | 1510/2562 [1:17:11<54:11,  3.09s/it]
                                                     
{'loss': 0.9112, 'grad_norm': 4.803908348083496, 'learning_rate': 7.891842714379027e-06, 'epoch': 1.77}

 59%|█████▉    | 1510/2562 [1:17:11<54:11,  3.09s/it]
 59%|█████▉    | 1511/2562 [1:17:13<52:39,  3.01s/it]
 59%|█████▉    | 1512/2562 [1:17:16<52:42,  3.01s/it]
 59%|█████▉    | 1513/2562 [1:17:19<52:09,  2.98s/it]
 59%|█████▉    | 1514/2562 [1:17:22<51:24,  2.94s/it]
 59%|█████▉    | 1515/2562 [1:17:25<52:16,  3.00s/it]
 59%|█████▉    | 1516/2562 [1:17:28<51:18,  2.94s/it]
 59%|█████▉    | 1517/2562 [1:17:31<50:33,  2.90s/it]
 59%|█████▉    | 1518/2562 [1:17:34<51:27,  2.96s/it]
 59%|█████▉    | 1519/2562 [1:17:37<50:50,  2.92s/it]
 59%|█████▉    | 1520/2562 [1:17:40<49:54,  2.87s/it]
                                                     
{'loss': 0.8938, 'grad_norm': 4.406307220458984, 'learning_rate': 7.765799692347201e-06, 'epoch': 1.78}

 59%|█████▉    | 1520/2562 [1:17:40<49:54,  2.87s/it]
 59%|█████▉    | 1521/2562 [1:17:43<50:58,  2.94s/it]
 59%|█████▉    | 1522/2562 [1:17:46<50:34,  2.92s/it]
 59%|█████▉    | 1523/2562 [1:17:49<51:06,  2.95s/it]
 59%|█████▉    | 1524/2562 [1:17:52<50:27,  2.92s/it]
 60%|█████▉    | 1525/2562 [1:17:54<49:37,  2.87s/it]
 60%|█████▉    | 1526/2562 [1:17:57<50:26,  2.92s/it]
 60%|█████▉    | 1527/2562 [1:18:00<49:48,  2.89s/it]
 60%|█████▉    | 1528/2562 [1:18:03<49:33,  2.88s/it]
 60%|█████▉    | 1529/2562 [1:18:06<49:39,  2.88s/it]
 60%|█████▉    | 1530/2562 [1:18:09<49:51,  2.90s/it]
                                                     
{'loss': 0.8765, 'grad_norm': 4.7871785163879395, 'learning_rate': 7.64012917488146e-06, 'epoch': 1.79}

 60%|█████▉    | 1530/2562 [1:18:09<49:51,  2.90s/it]
 60%|█████▉    | 1531/2562 [1:18:12<49:04,  2.86s/it]
 60%|█████▉    | 1532/2562 [1:18:14<49:09,  2.86s/it]
 60%|█████▉    | 1533/2562 [1:18:17<48:51,  2.85s/it]
 60%|█████▉    | 1534/2562 [1:18:20<48:19,  2.82s/it]
 60%|█████▉    | 1535/2562 [1:18:23<49:17,  2.88s/it]
 60%|█████▉    | 1536/2562 [1:18:26<49:34,  2.90s/it]
 60%|█████▉    | 1537/2562 [1:18:29<49:09,  2.88s/it]
 60%|██████    | 1538/2562 [1:18:32<50:06,  2.94s/it]
 60%|██████    | 1539/2562 [1:18:35<49:26,  2.90s/it]
 60%|██████    | 1540/2562 [1:18:38<49:31,  2.91s/it]
                                                     
{'loss': 0.887, 'grad_norm': 4.98879337310791, 'learning_rate': 7.514852114821811e-06, 'epoch': 1.8}

 60%|██████    | 1540/2562 [1:18:38<49:31,  2.91s/it]
 60%|██████    | 1541/2562 [1:18:41<49:56,  2.94s/it]
 60%|██████    | 1542/2562 [1:18:43<49:25,  2.91s/it]
 60%|██████    | 1543/2562 [1:18:46<49:33,  2.92s/it]
 60%|██████    | 1544/2562 [1:18:50<50:42,  2.99s/it]
 60%|██████    | 1545/2562 [1:18:52<49:25,  2.92s/it]
 60%|██████    | 1546/2562 [1:18:56<51:22,  3.03s/it]
 60%|██████    | 1547/2562 [1:18:58<50:14,  2.97s/it]
 60%|██████    | 1548/2562 [1:19:01<49:58,  2.96s/it]
 60%|██████    | 1549/2562 [1:19:04<49:59,  2.96s/it]
 60%|██████    | 1550/2562 [1:19:07<50:01,  2.97s/it]
                                                     
{'loss': 0.8867, 'grad_norm': 4.977054595947266, 'learning_rate': 7.389989399407741e-06, 'epoch': 1.81}

 60%|██████    | 1550/2562 [1:19:07<50:01,  2.97s/it]
 61%|██████    | 1551/2562 [1:19:10<49:31,  2.94s/it]
 61%|██████    | 1552/2562 [1:19:13<49:03,  2.91s/it]
 61%|██████    | 1553/2562 [1:19:16<48:46,  2.90s/it]
 61%|██████    | 1554/2562 [1:19:19<49:18,  2.94s/it]
 61%|██████    | 1555/2562 [1:19:22<50:09,  2.99s/it]
 61%|██████    | 1556/2562 [1:19:25<50:13,  3.00s/it]
 61%|██████    | 1557/2562 [1:19:28<49:07,  2.93s/it]
 61%|██████    | 1558/2562 [1:19:31<49:10,  2.94s/it]
 61%|██████    | 1559/2562 [1:19:34<48:18,  2.89s/it]
 61%|██████    | 1560/2562 [1:19:37<48:45,  2.92s/it]
                                                     
{'loss': 0.8529, 'grad_norm': 4.643401145935059, 'learning_rate': 7.265561846795741e-06, 'epoch': 1.83}

 61%|██████    | 1560/2562 [1:19:37<48:45,  2.92s/it]
 61%|██████    | 1561/2562 [1:19:39<48:24,  2.90s/it]
 61%|██████    | 1562/2562 [1:19:42<48:03,  2.88s/it]
 61%|██████    | 1563/2562 [1:19:45<47:53,  2.88s/it]
 61%|██████    | 1564/2562 [1:19:48<48:07,  2.89s/it]
 61%|██████    | 1565/2562 [1:19:51<47:47,  2.88s/it]
 61%|██████    | 1566/2562 [1:19:54<47:14,  2.85s/it]
 61%|██████    | 1567/2562 [1:19:57<47:42,  2.88s/it]
 61%|██████    | 1568/2562 [1:20:00<48:00,  2.90s/it]
 61%|██████    | 1569/2562 [1:20:02<47:43,  2.88s/it]
 61%|██████▏   | 1570/2562 [1:20:05<47:56,  2.90s/it]
                                                     
{'loss': 0.8789, 'grad_norm': 5.361824989318848, 'learning_rate': 7.141590202588312e-06, 'epoch': 1.84}

 61%|██████▏   | 1570/2562 [1:20:05<47:56,  2.90s/it]
 61%|██████▏   | 1571/2562 [1:20:08<47:34,  2.88s/it]
 61%|██████▏   | 1572/2562 [1:20:11<48:07,  2.92s/it]
 61%|██████▏   | 1573/2562 [1:20:14<47:58,  2.91s/it]
 61%|██████▏   | 1574/2562 [1:20:17<47:52,  2.91s/it]
 61%|██████▏   | 1575/2562 [1:20:20<47:59,  2.92s/it]
 62%|██████▏   | 1576/2562 [1:20:23<47:32,  2.89s/it]
 62%|██████▏   | 1577/2562 [1:20:26<46:55,  2.86s/it]
 62%|██████▏   | 1578/2562 [1:20:29<48:10,  2.94s/it]
 62%|██████▏   | 1579/2562 [1:20:31<47:15,  2.89s/it]
 62%|██████▏   | 1580/2562 [1:20:34<47:27,  2.90s/it]
                                                     
{'loss': 0.871, 'grad_norm': 4.649825096130371, 'learning_rate': 7.018095136375089e-06, 'epoch': 1.85}

 62%|██████▏   | 1580/2562 [1:20:34<47:27,  2.90s/it]
 62%|██████▏   | 1581/2562 [1:20:37<48:29,  2.97s/it]
 62%|██████▏   | 1582/2562 [1:20:40<47:46,  2.93s/it]
 62%|██████▏   | 1583/2562 [1:20:43<46:55,  2.88s/it]
 62%|██████▏   | 1584/2562 [1:20:46<47:37,  2.92s/it]
 62%|██████▏   | 1585/2562 [1:20:49<47:32,  2.92s/it]
 62%|██████▏   | 1586/2562 [1:20:52<46:59,  2.89s/it]
 62%|██████▏   | 1587/2562 [1:20:55<47:12,  2.91s/it]
 62%|██████▏   | 1588/2562 [1:20:58<46:27,  2.86s/it]
 62%|██████▏   | 1589/2562 [1:21:00<46:34,  2.87s/it]
 62%|██████▏   | 1590/2562 [1:21:03<46:47,  2.89s/it]
                                                     
{'loss': 0.8782, 'grad_norm': 4.606133937835693, 'learning_rate': 6.89509723828665e-06, 'epoch': 1.86}

 62%|██████▏   | 1590/2562 [1:21:03<46:47,  2.89s/it]
 62%|██████▏   | 1591/2562 [1:21:06<47:17,  2.92s/it]
 62%|██████▏   | 1592/2562 [1:21:09<48:18,  2.99s/it]
 62%|██████▏   | 1593/2562 [1:21:12<47:39,  2.95s/it]
 62%|██████▏   | 1594/2562 [1:21:15<46:36,  2.89s/it]
 62%|██████▏   | 1595/2562 [1:21:18<47:11,  2.93s/it]
 62%|██████▏   | 1596/2562 [1:21:21<46:44,  2.90s/it]
 62%|██████▏   | 1597/2562 [1:21:24<46:04,  2.86s/it]
 62%|██████▏   | 1598/2562 [1:21:27<46:38,  2.90s/it]
 62%|██████▏   | 1599/2562 [1:21:30<46:01,  2.87s/it]
 62%|██████▏   | 1600/2562 [1:21:32<45:44,  2.85s/it]
                                                     
{'loss': 0.8881, 'grad_norm': 4.930484294891357, 'learning_rate': 6.772617015561529e-06, 'epoch': 1.87}

 62%|██████▏   | 1600/2562 [1:21:32<45:44,  2.85s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.66it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.74it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.21it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.89it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.55it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.74it/s]

 10%|█         | 14/140 [00:01<00:12,  9.78it/s]

 11%|█         | 15/140 [00:01<00:12,  9.81it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.82it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.86it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.65it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.59it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.69it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.44it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.40it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.54it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.39it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.08it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.16it/s]

 20%|██        | 28/140 [00:02<00:11,  9.50it/s]

 21%|██        | 29/140 [00:02<00:11,  9.35it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.04it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.05it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.28it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.74it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.82it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.89it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.96it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.96it/s]

 31%|███       | 43/140 [00:04<00:09, 10.00it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.96it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.93it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.67it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.34it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.30it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.21it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.21it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.42it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.42it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.42it/s]

 40%|████      | 56/140 [00:05<00:08,  9.69it/s]

 41%|████      | 57/140 [00:05<00:08,  9.49it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.41it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.52it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.63it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.85it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.65it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.58it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.55it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.53it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.72it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.81it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.85it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.87it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.87it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.86it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.61it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.23it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.38it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.31it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.22it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.13it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.13it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.34it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.22it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.21it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.10it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.09it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.00it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.10it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.30it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.45it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.30it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.05it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.15it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.29it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.33it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.46it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.28it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.34it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.68it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.71it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.64it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.53it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.60it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.66it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.72it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.88it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.87it/s]

 82%|████████▏ | 115/140 [00:11<00:02,  9.53it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.44it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.56it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.32it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.27it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.11it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.07it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.22it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.39it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.29it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.06it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.85it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.57it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.69it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.02it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.49it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.61it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.62it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.44it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.49it/s]

100%|██████████| 140/140 [00:14<00:00,  8.05it/s]
                                                     

                                                 
{'eval_loss': 0.941744327545166, 'eval_runtime': 15.1068, 'eval_samples_per_second': 18.535, 'eval_steps_per_second': 9.267, 'epoch': 1.87}

 62%|██████▏   | 1600/2562 [1:21:47<45:44,  2.85s/it]

100%|██████████| 140/140 [00:14<00:00,  8.05it/s]

                                                 
 62%|██████▏   | 1601/2562 [1:21:50<1:58:47,  7.42s/it]
 63%|██████▎   | 1602/2562 [1:21:53<1:36:07,  6.01s/it]
 63%|██████▎   | 1603/2562 [1:21:56<1:21:14,  5.08s/it]
 63%|██████▎   | 1604/2562 [1:21:59<1:10:40,  4.43s/it]
 63%|██████▎   | 1605/2562 [1:22:02<1:03:32,  3.98s/it]
 63%|██████▎   | 1606/2562 [1:22:05<57:49,  3.63s/it]  
 63%|██████▎   | 1607/2562 [1:22:08<55:05,  3.46s/it]
 63%|██████▎   | 1608/2562 [1:22:11<51:52,  3.26s/it]
 63%|██████▎   | 1609/2562 [1:22:13<49:47,  3.13s/it]
 63%|██████▎   | 1610/2562 [1:22:16<48:14,  3.04s/it]
                                                     
{'loss': 0.8983, 'grad_norm': 5.305909156799316, 'learning_rate': 6.6506748891271045e-06, 'epoch': 1.88}

 63%|██████▎   | 1610/2562 [1:22:16<48:14,  3.04s/it]
 63%|██████▎   | 1611/2562 [1:22:19<47:12,  2.98s/it]
 63%|██████▎   | 1612/2562 [1:22:22<47:04,  2.97s/it]
 63%|██████▎   | 1613/2562 [1:22:25<46:52,  2.96s/it]
 63%|██████▎   | 1614/2562 [1:22:28<46:37,  2.95s/it]
 63%|██████▎   | 1615/2562 [1:22:31<45:37,  2.89s/it]
 63%|██████▎   | 1616/2562 [1:22:34<45:52,  2.91s/it]
 63%|██████▎   | 1617/2562 [1:22:36<45:26,  2.88s/it]
 63%|██████▎   | 1618/2562 [1:22:39<45:35,  2.90s/it]
 63%|██████▎   | 1619/2562 [1:22:42<45:47,  2.91s/it]
 63%|██████▎   | 1620/2562 [1:22:45<45:52,  2.92s/it]
                                                     
{'loss': 0.8641, 'grad_norm': 4.630437850952148, 'learning_rate': 6.529291190194829e-06, 'epoch': 1.9}

 63%|██████▎   | 1620/2562 [1:22:45<45:52,  2.92s/it]
 63%|██████▎   | 1621/2562 [1:22:48<44:54,  2.86s/it]
 63%|██████▎   | 1622/2562 [1:22:51<45:26,  2.90s/it]
 63%|██████▎   | 1623/2562 [1:22:54<45:43,  2.92s/it]
 63%|██████▎   | 1624/2562 [1:22:57<45:57,  2.94s/it]
 63%|██████▎   | 1625/2562 [1:23:00<45:12,  2.89s/it]
 63%|██████▎   | 1626/2562 [1:23:03<44:54,  2.88s/it]
 64%|██████▎   | 1627/2562 [1:23:06<45:33,  2.92s/it]
 64%|██████▎   | 1628/2562 [1:23:08<44:36,  2.87s/it]
 64%|██████▎   | 1629/2562 [1:23:11<43:38,  2.81s/it]
 64%|██████▎   | 1630/2562 [1:23:14<44:37,  2.87s/it]
                                                     
{'loss': 0.8877, 'grad_norm': 4.860596179962158, 'learning_rate': 6.408486156870466e-06, 'epoch': 1.91}

 64%|██████▎   | 1630/2562 [1:23:14<44:37,  2.87s/it]
 64%|██████▎   | 1631/2562 [1:23:17<44:40,  2.88s/it]
 64%|██████▎   | 1632/2562 [1:23:20<44:44,  2.89s/it]
 64%|██████▎   | 1633/2562 [1:23:23<44:55,  2.90s/it]
 64%|██████▍   | 1634/2562 [1:23:25<44:02,  2.85s/it]
 64%|██████▍   | 1635/2562 [1:23:28<43:17,  2.80s/it]
 64%|██████▍   | 1636/2562 [1:23:31<44:17,  2.87s/it]
 64%|██████▍   | 1637/2562 [1:23:34<44:09,  2.86s/it]
 64%|██████▍   | 1638/2562 [1:23:37<44:14,  2.87s/it]
 64%|██████▍   | 1639/2562 [1:23:40<44:27,  2.89s/it]
 64%|██████▍   | 1640/2562 [1:23:43<43:50,  2.85s/it]
                                                     
{'loss': 0.8833, 'grad_norm': 5.3686113357543945, 'learning_rate': 6.288279930779789e-06, 'epoch': 1.92}

 64%|██████▍   | 1640/2562 [1:23:43<43:50,  2.85s/it]
 64%|██████▍   | 1641/2562 [1:23:45<43:56,  2.86s/it]
 64%|██████▍   | 1642/2562 [1:23:49<44:35,  2.91s/it]
 64%|██████▍   | 1643/2562 [1:23:51<44:14,  2.89s/it]
 64%|██████▍   | 1644/2562 [1:23:54<43:51,  2.87s/it]
 64%|██████▍   | 1645/2562 [1:23:57<43:57,  2.88s/it]
 64%|██████▍   | 1646/2562 [1:24:00<43:36,  2.86s/it]
 64%|██████▍   | 1647/2562 [1:24:03<43:25,  2.85s/it]
 64%|██████▍   | 1648/2562 [1:24:06<44:22,  2.91s/it]
 64%|██████▍   | 1649/2562 [1:24:09<43:52,  2.88s/it]
 64%|██████▍   | 1650/2562 [1:24:12<44:02,  2.90s/it]
                                                     
{'loss': 0.866, 'grad_norm': 4.888591766357422, 'learning_rate': 6.1686925537104306e-06, 'epoch': 1.93}

 64%|██████▍   | 1650/2562 [1:24:12<44:02,  2.90s/it]
 64%|██████▍   | 1651/2562 [1:24:14<44:12,  2.91s/it]
 64%|██████▍   | 1652/2562 [1:24:17<44:40,  2.95s/it]
 65%|██████▍   | 1653/2562 [1:24:20<44:52,  2.96s/it]
 65%|██████▍   | 1654/2562 [1:24:23<44:51,  2.96s/it]
 65%|██████▍   | 1655/2562 [1:24:26<44:03,  2.91s/it]
 65%|██████▍   | 1656/2562 [1:24:29<43:11,  2.86s/it]
 65%|██████▍   | 1657/2562 [1:24:32<43:34,  2.89s/it]
 65%|██████▍   | 1658/2562 [1:24:35<43:21,  2.88s/it]
 65%|██████▍   | 1659/2562 [1:24:38<42:40,  2.84s/it]
 65%|██████▍   | 1660/2562 [1:24:40<42:53,  2.85s/it]
                                                     
{'loss': 0.8651, 'grad_norm': 4.506246089935303, 'learning_rate': 6.049743964270336e-06, 'epoch': 1.94}

 65%|██████▍   | 1660/2562 [1:24:40<42:53,  2.85s/it]
 65%|██████▍   | 1661/2562 [1:24:43<43:36,  2.90s/it]
 65%|██████▍   | 1662/2562 [1:24:46<43:05,  2.87s/it]
 65%|██████▍   | 1663/2562 [1:24:49<43:31,  2.91s/it]
 65%|██████▍   | 1664/2562 [1:24:52<43:00,  2.87s/it]
 65%|██████▍   | 1665/2562 [1:24:55<43:09,  2.89s/it]
 65%|██████▌   | 1666/2562 [1:24:58<43:37,  2.92s/it]
 65%|██████▌   | 1667/2562 [1:25:01<43:13,  2.90s/it]
 65%|██████▌   | 1668/2562 [1:25:04<43:47,  2.94s/it]
 65%|██████▌   | 1669/2562 [1:25:07<43:44,  2.94s/it]
 65%|██████▌   | 1670/2562 [1:25:10<43:24,  2.92s/it]
                                                     
{'loss': 0.8692, 'grad_norm': 4.978627681732178, 'learning_rate': 5.931453994563434e-06, 'epoch': 1.95}

 65%|██████▌   | 1670/2562 [1:25:10<43:24,  2.92s/it]
 65%|██████▌   | 1671/2562 [1:25:13<43:29,  2.93s/it]
 65%|██████▌   | 1672/2562 [1:25:15<43:16,  2.92s/it]
 65%|██████▌   | 1673/2562 [1:25:18<42:46,  2.89s/it]
 65%|██████▌   | 1674/2562 [1:25:21<43:51,  2.96s/it]
 65%|██████▌   | 1675/2562 [1:25:24<43:10,  2.92s/it]
 65%|██████▌   | 1676/2562 [1:25:27<42:24,  2.87s/it]
 65%|██████▌   | 1677/2562 [1:25:30<43:16,  2.93s/it]
 65%|██████▌   | 1678/2562 [1:25:33<43:02,  2.92s/it]
 66%|██████▌   | 1679/2562 [1:25:36<42:19,  2.88s/it]
 66%|██████▌   | 1680/2562 [1:25:39<43:07,  2.93s/it]
                                                     
{'loss': 0.8534, 'grad_norm': 4.72692346572876, 'learning_rate': 5.8138423668830605e-06, 'epoch': 1.97}

 66%|██████▌   | 1680/2562 [1:25:39<43:07,  2.93s/it]
 66%|██████▌   | 1681/2562 [1:25:42<42:34,  2.90s/it]
 66%|██████▌   | 1682/2562 [1:25:45<42:24,  2.89s/it]
 66%|██████▌   | 1683/2562 [1:25:47<42:13,  2.88s/it]
 66%|██████▌   | 1684/2562 [1:25:50<41:45,  2.85s/it]
 66%|██████▌   | 1685/2562 [1:25:53<42:06,  2.88s/it]
 66%|██████▌   | 1686/2562 [1:25:56<42:59,  2.94s/it]
 66%|██████▌   | 1687/2562 [1:25:59<42:18,  2.90s/it]
 66%|██████▌   | 1688/2562 [1:26:02<42:04,  2.89s/it]
 66%|██████▌   | 1689/2562 [1:26:05<42:02,  2.89s/it]
 66%|██████▌   | 1690/2562 [1:26:08<42:13,  2.91s/it]
                                                     
{'loss': 0.8591, 'grad_norm': 4.491663932800293, 'learning_rate': 5.696928690423693e-06, 'epoch': 1.98}

 66%|██████▌   | 1690/2562 [1:26:08<42:13,  2.91s/it]
 66%|██████▌   | 1691/2562 [1:26:10<41:31,  2.86s/it]
 66%|██████▌   | 1692/2562 [1:26:13<41:47,  2.88s/it]
 66%|██████▌   | 1693/2562 [1:26:16<41:29,  2.87s/it]
 66%|██████▌   | 1694/2562 [1:26:19<41:06,  2.84s/it]
 66%|██████▌   | 1695/2562 [1:26:22<41:34,  2.88s/it]
 66%|██████▌   | 1696/2562 [1:26:25<41:10,  2.85s/it]
 66%|██████▌   | 1697/2562 [1:26:28<40:48,  2.83s/it]
 66%|██████▋   | 1698/2562 [1:26:31<41:30,  2.88s/it]
 66%|██████▋   | 1699/2562 [1:26:34<41:53,  2.91s/it]
 66%|██████▋   | 1700/2562 [1:26:36<41:34,  2.89s/it]
                                                     
{'loss': 0.8494, 'grad_norm': 4.865417003631592, 'learning_rate': 5.580732458011544e-06, 'epoch': 1.99}

 66%|██████▋   | 1700/2562 [1:26:36<41:34,  2.89s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.76it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.72it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.21it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.88it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.55it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.80it/s]

 11%|█         | 15/140 [00:01<00:12,  9.83it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.81it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.84it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.63it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.56it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.66it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.45it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.40it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.53it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.39it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.10it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.16it/s]

 20%|██        | 28/140 [00:02<00:11,  9.46it/s]

 21%|██        | 29/140 [00:02<00:11,  9.31it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.01it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.02it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.24it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.72it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.84it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.85it/s]

 28%|██▊       | 39/140 [00:03<00:10,  9.95it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.96it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.90it/s]

 30%|███       | 42/140 [00:04<00:09,  9.92it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.93it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.85it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.62it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.30it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.27it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.19it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.24it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.44it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.45it/s]

 40%|████      | 56/140 [00:05<00:08,  9.73it/s]

 41%|████      | 57/140 [00:05<00:08,  9.51it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.42it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.52it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.61it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.82it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.62it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.55it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.51it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.50it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.73it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.79it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.80it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.85it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.88it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.88it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.87it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.63it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.24it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.39it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.32it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.23it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.11it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.13it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.32it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.21it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.13it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.08it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.09it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.99it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.07it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.25it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.42it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.25it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.02it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.10it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.32it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.23it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.31it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.44it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.27it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.36it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.71it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.73it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.61it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.50it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.59it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.67it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.84it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.84it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.54it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.43it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.56it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.32it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.30it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.12it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.08it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.24it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.21it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.36it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.25it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.05it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.83it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.60it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.72it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.03it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.28it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.67it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.67it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.54it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.55it/s]

100%|██████████| 140/140 [00:14<00:00,  8.10it/s]
                                                     

                                                 
{'eval_loss': 0.9274364113807678, 'eval_runtime': 15.1183, 'eval_samples_per_second': 18.521, 'eval_steps_per_second': 9.26, 'epoch': 1.99}

 66%|██████▋   | 1700/2562 [1:26:51<41:34,  2.89s/it]

100%|██████████| 140/140 [00:14<00:00,  8.10it/s]

                                                 
 66%|██████▋   | 1701/2562 [1:26:55<1:47:19,  7.48s/it]
 66%|██████▋   | 1702/2562 [1:26:57<1:27:13,  6.09s/it]
 66%|██████▋   | 1703/2562 [1:27:00<1:13:49,  5.16s/it]
 67%|██████▋   | 1704/2562 [1:27:03<1:03:13,  4.42s/it]
 67%|██████▋   | 1705/2562 [1:27:06<56:28,  3.95s/it]  
 67%|██████▋   | 1706/2562 [1:27:09<52:10,  3.66s/it]
 67%|██████▋   | 1707/2562 [1:27:12<47:59,  3.37s/it]
 67%|██████▋   | 1708/2562 [1:27:15<46:10,  3.24s/it]
 67%|██████▋   | 1709/2562 [1:27:19<53:18,  3.75s/it]
 67%|██████▋   | 1710/2562 [1:27:22<49:37,  3.49s/it]
                                                     
{'loss': 0.9592, 'grad_norm': 4.482592582702637, 'learning_rate': 5.465273042854551e-06, 'epoch': 2.0}

 67%|██████▋   | 1710/2562 [1:27:22<49:37,  3.49s/it]
 67%|██████▋   | 1711/2562 [1:27:25<47:33,  3.35s/it]
 67%|██████▋   | 1712/2562 [1:27:28<45:40,  3.22s/it]
 67%|██████▋   | 1713/2562 [1:27:31<43:37,  3.08s/it]
 67%|██████▋   | 1714/2562 [1:27:34<43:00,  3.04s/it]
 67%|██████▋   | 1715/2562 [1:27:37<41:25,  2.93s/it]
 67%|██████▋   | 1716/2562 [1:27:40<41:27,  2.94s/it]
 67%|██████▋   | 1717/2562 [1:27:42<40:37,  2.88s/it]
 67%|██████▋   | 1718/2562 [1:27:45<40:18,  2.87s/it]
 67%|██████▋   | 1719/2562 [1:27:48<40:01,  2.85s/it]
 67%|██████▋   | 1720/2562 [1:27:51<40:23,  2.88s/it]
                                                     
{'loss': 0.8055, 'grad_norm': 4.645415306091309, 'learning_rate': 5.350569695312313e-06, 'epoch': 2.01}

 67%|██████▋   | 1720/2562 [1:27:51<40:23,  2.88s/it]
 67%|██████▋   | 1721/2562 [1:27:54<40:08,  2.86s/it]
 67%|██████▋   | 1722/2562 [1:27:57<39:56,  2.85s/it]
 67%|██████▋   | 1723/2562 [1:28:00<41:12,  2.95s/it]
 67%|██████▋   | 1724/2562 [1:28:03<41:20,  2.96s/it]
 67%|██████▋   | 1725/2562 [1:28:06<40:47,  2.92s/it]
 67%|██████▋   | 1726/2562 [1:28:09<40:52,  2.93s/it]
 67%|██████▋   | 1727/2562 [1:28:11<40:18,  2.90s/it]
 67%|██████▋   | 1728/2562 [1:28:14<40:14,  2.89s/it]
 67%|██████▋   | 1729/2562 [1:28:17<41:00,  2.95s/it]
 68%|██████▊   | 1730/2562 [1:28:20<40:04,  2.89s/it]
                                                     
{'loss': 0.824, 'grad_norm': 4.703752517700195, 'learning_rate': 5.236641539686518e-06, 'epoch': 2.03}

 68%|██████▊   | 1730/2562 [1:28:20<40:04,  2.89s/it]
 68%|██████▊   | 1731/2562 [1:28:23<39:06,  2.82s/it]
 68%|██████▊   | 1732/2562 [1:28:26<40:10,  2.90s/it]
 68%|██████▊   | 1733/2562 [1:28:29<40:19,  2.92s/it]
 68%|██████▊   | 1734/2562 [1:28:32<40:38,  2.95s/it]
 68%|██████▊   | 1735/2562 [1:28:35<40:50,  2.96s/it]
 68%|██████▊   | 1736/2562 [1:28:38<40:38,  2.95s/it]
 68%|██████▊   | 1737/2562 [1:28:41<40:31,  2.95s/it]
 68%|██████▊   | 1738/2562 [1:28:44<40:11,  2.93s/it]
 68%|██████▊   | 1739/2562 [1:28:47<40:06,  2.92s/it]
 68%|██████▊   | 1740/2562 [1:28:50<40:44,  2.97s/it]
                                                     
{'loss': 0.7769, 'grad_norm': 4.4950385093688965, 'learning_rate': 5.12350757103236e-06, 'epoch': 2.04}

 68%|██████▊   | 1740/2562 [1:28:50<40:44,  2.97s/it]
 68%|██████▊   | 1741/2562 [1:28:53<40:48,  2.98s/it]
 68%|██████▊   | 1742/2562 [1:28:56<40:29,  2.96s/it]
 68%|██████▊   | 1743/2562 [1:28:58<39:59,  2.93s/it]
 68%|██████▊   | 1744/2562 [1:29:02<41:01,  3.01s/it]
 68%|██████▊   | 1745/2562 [1:29:04<40:24,  2.97s/it]
 68%|██████▊   | 1746/2562 [1:29:07<39:42,  2.92s/it]
 68%|██████▊   | 1747/2562 [1:29:10<39:51,  2.93s/it]
 68%|██████▊   | 1748/2562 [1:29:13<39:24,  2.90s/it]
 68%|██████▊   | 1749/2562 [1:29:16<39:14,  2.90s/it]
 68%|██████▊   | 1750/2562 [1:29:19<39:13,  2.90s/it]
                                                     
{'loss': 0.7878, 'grad_norm': 4.889289855957031, 'learning_rate': 5.0111866519915575e-06, 'epoch': 2.05}

 68%|██████▊   | 1750/2562 [1:29:19<39:13,  2.90s/it]
 68%|██████▊   | 1751/2562 [1:29:22<39:02,  2.89s/it]
 68%|██████▊   | 1752/2562 [1:29:25<39:31,  2.93s/it]
 68%|██████▊   | 1753/2562 [1:29:28<39:29,  2.93s/it]
 68%|██████▊   | 1754/2562 [1:29:31<39:21,  2.92s/it]
 69%|██████▊   | 1755/2562 [1:29:33<38:35,  2.87s/it]
 69%|██████▊   | 1756/2562 [1:29:36<38:55,  2.90s/it]
 69%|██████▊   | 1757/2562 [1:29:39<38:33,  2.87s/it]
 69%|██████▊   | 1758/2562 [1:29:42<38:40,  2.89s/it]
 69%|██████▊   | 1759/2562 [1:29:45<38:39,  2.89s/it]
 69%|██████▊   | 1760/2562 [1:29:48<38:33,  2.88s/it]
                                                     
{'loss': 0.7879, 'grad_norm': 4.696087837219238, 'learning_rate': 4.899697509647379e-06, 'epoch': 2.06}

 69%|██████▊   | 1760/2562 [1:29:48<38:33,  2.88s/it]
 69%|██████▊   | 1761/2562 [1:29:51<38:53,  2.91s/it]
 69%|██████▉   | 1762/2562 [1:29:54<38:39,  2.90s/it]
 69%|██████▉   | 1763/2562 [1:29:57<38:35,  2.90s/it]
 69%|██████▉   | 1764/2562 [1:29:59<38:37,  2.90s/it]
 69%|██████▉   | 1765/2562 [1:30:02<38:25,  2.89s/it]
 69%|██████▉   | 1766/2562 [1:30:05<38:08,  2.87s/it]
 69%|██████▉   | 1767/2562 [1:30:08<37:46,  2.85s/it]
 69%|██████▉   | 1768/2562 [1:30:11<38:10,  2.88s/it]
 69%|██████▉   | 1769/2562 [1:30:14<38:27,  2.91s/it]
 69%|██████▉   | 1770/2562 [1:30:17<39:13,  2.97s/it]
                                                     
{'loss': 0.806, 'grad_norm': 4.908601760864258, 'learning_rate': 4.789058732402319e-06, 'epoch': 2.07}

 69%|██████▉   | 1770/2562 [1:30:17<39:13,  2.97s/it]
 69%|██████▉   | 1771/2562 [1:30:20<38:15,  2.90s/it]
 69%|██████▉   | 1772/2562 [1:30:23<37:49,  2.87s/it]
 69%|██████▉   | 1773/2562 [1:30:25<37:31,  2.85s/it]
 69%|██████▉   | 1774/2562 [1:30:28<37:54,  2.89s/it]
 69%|██████▉   | 1775/2562 [1:30:31<38:20,  2.92s/it]
 69%|██████▉   | 1776/2562 [1:30:34<38:50,  2.96s/it]
 69%|██████▉   | 1777/2562 [1:30:37<37:43,  2.88s/it]
 69%|██████▉   | 1778/2562 [1:30:40<37:35,  2.88s/it]
 69%|██████▉   | 1779/2562 [1:30:43<37:47,  2.90s/it]
 69%|██████▉   | 1780/2562 [1:30:46<37:39,  2.89s/it]
                                                     
{'loss': 0.8195, 'grad_norm': 4.76207160949707, 'learning_rate': 4.679288766878908e-06, 'epoch': 2.08}

 69%|██████▉   | 1780/2562 [1:30:46<37:39,  2.89s/it]
 70%|██████▉   | 1781/2562 [1:30:48<37:07,  2.85s/it]
 70%|██████▉   | 1782/2562 [1:30:52<38:29,  2.96s/it]
 70%|██████▉   | 1783/2562 [1:30:55<37:52,  2.92s/it]
 70%|██████▉   | 1784/2562 [1:30:58<38:03,  2.94s/it]
 70%|██████▉   | 1785/2562 [1:31:01<38:22,  2.96s/it]
 70%|██████▉   | 1786/2562 [1:31:03<37:43,  2.92s/it]
 70%|██████▉   | 1787/2562 [1:31:06<38:03,  2.95s/it]
 70%|██████▉   | 1788/2562 [1:31:09<37:11,  2.88s/it]
 70%|██████▉   | 1789/2562 [1:31:12<37:13,  2.89s/it]
 70%|██████▉   | 1790/2562 [1:31:15<37:22,  2.90s/it]
                                                     
{'loss': 0.8022, 'grad_norm': 4.682501316070557, 'learning_rate': 4.570405914844105e-06, 'epoch': 2.1}

 70%|██████▉   | 1790/2562 [1:31:15<37:22,  2.90s/it]
 70%|██████▉   | 1791/2562 [1:31:18<37:26,  2.91s/it]
 70%|██████▉   | 1792/2562 [1:31:21<37:12,  2.90s/it]
 70%|██████▉   | 1793/2562 [1:31:24<37:21,  2.91s/it]
 70%|███████   | 1794/2562 [1:31:27<37:02,  2.89s/it]
 70%|███████   | 1795/2562 [1:31:29<36:29,  2.85s/it]
 70%|███████   | 1796/2562 [1:31:32<37:17,  2.92s/it]
 70%|███████   | 1797/2562 [1:31:35<36:53,  2.89s/it]
 70%|███████   | 1798/2562 [1:31:38<36:39,  2.88s/it]
 70%|███████   | 1799/2562 [1:31:41<36:51,  2.90s/it]
 70%|███████   | 1800/2562 [1:31:44<36:29,  2.87s/it]
                                                     
{'loss': 0.8183, 'grad_norm': 4.834522724151611, 'learning_rate': 4.462428330157886e-06, 'epoch': 2.11}

 70%|███████   | 1800/2562 [1:31:44<36:29,  2.87s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.58it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.74it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.23it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.88it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.54it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.78it/s]

 11%|█         | 15/140 [00:01<00:12,  9.80it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.79it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.83it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.62it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.56it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.67it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.44it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.39it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.52it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.36it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.06it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.13it/s]

 20%|██        | 28/140 [00:02<00:11,  9.48it/s]

 21%|██        | 29/140 [00:02<00:11,  9.32it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.03it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.04it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.24it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.69it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.82it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.85it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.94it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.93it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.93it/s]

 31%|███       | 43/140 [00:04<00:09,  9.97it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.93it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.91it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.67it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.31it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.26it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.16it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.24it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.45it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.47it/s]

 40%|████      | 56/140 [00:05<00:08,  9.73it/s]

 41%|████      | 57/140 [00:05<00:08,  9.51it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.41it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.58it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.82it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.64it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.58it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.53it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.52it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.74it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.81it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.82it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.84it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.85it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.84it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.85it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.60it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.22it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.35it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.32it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.23it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.14it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.12it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.33it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.18it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.15it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.07it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.09it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.99it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.07it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.28it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.45it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.28it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.02it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.13it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.37it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.27it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.33it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.52it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.31it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.35it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.74it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.74it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.60it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.47it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.57it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.62it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.64it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.82it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.83it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.52it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.41it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.52it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.32it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.42it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.13it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.09it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.26it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.24it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.42it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.31it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.10it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.86it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.61it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.75it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.06it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.51it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.66it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.51it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.53it/s]

100%|██████████| 140/140 [00:14<00:00,  8.11it/s]
                                                     

                                                 
{'eval_loss': 0.9262279868125916, 'eval_runtime': 15.1239, 'eval_samples_per_second': 18.514, 'eval_steps_per_second': 9.257, 'epoch': 2.11}

 70%|███████   | 1800/2562 [1:31:59<36:29,  2.87s/it]

100%|██████████| 140/140 [00:14<00:00,  8.11it/s]

                                                 
 70%|███████   | 1801/2562 [1:32:02<1:34:13,  7.43s/it]
 70%|███████   | 1802/2562 [1:32:05<1:17:16,  6.10s/it]
 70%|███████   | 1803/2562 [1:32:08<1:04:34,  5.10s/it]
 70%|███████   | 1804/2562 [1:32:10<55:46,  4.42s/it]  
 70%|███████   | 1805/2562 [1:32:13<50:29,  4.00s/it]
 70%|███████   | 1806/2562 [1:32:16<45:46,  3.63s/it]
 71%|███████   | 1807/2562 [1:32:19<43:05,  3.42s/it]
 71%|███████   | 1808/2562 [1:32:22<41:50,  3.33s/it]
 71%|███████   | 1809/2562 [1:32:25<40:27,  3.22s/it]
 71%|███████   | 1810/2562 [1:32:29<40:21,  3.22s/it]
                                                     
{'loss': 0.8316, 'grad_norm': 4.9211225509643555, 'learning_rate': 4.355374015746493e-06, 'epoch': 2.12}

 71%|███████   | 1810/2562 [1:32:29<40:21,  3.22s/it]
 71%|███████   | 1811/2562 [1:32:31<39:02,  3.12s/it]
 71%|███████   | 1812/2562 [1:32:34<37:55,  3.03s/it]
 71%|███████   | 1813/2562 [1:32:37<37:38,  3.02s/it]
 71%|███████   | 1814/2562 [1:32:40<36:42,  2.94s/it]
 71%|███████   | 1815/2562 [1:32:43<36:44,  2.95s/it]
 71%|███████   | 1816/2562 [1:32:46<37:03,  2.98s/it]
 71%|███████   | 1817/2562 [1:32:49<36:18,  2.92s/it]
 71%|███████   | 1818/2562 [1:32:52<35:46,  2.88s/it]
 71%|███████   | 1819/2562 [1:32:55<36:20,  2.94s/it]
 71%|███████   | 1820/2562 [1:32:57<35:33,  2.88s/it]
                                                     
{'loss': 0.8032, 'grad_norm': 4.872379779815674, 'learning_rate': 4.249260820600813e-06, 'epoch': 2.13}

 71%|███████   | 1820/2562 [1:32:57<35:33,  2.88s/it]
 71%|███████   | 1821/2562 [1:33:00<35:42,  2.89s/it]
 71%|███████   | 1822/2562 [1:33:03<36:18,  2.94s/it]
 71%|███████   | 1823/2562 [1:33:06<35:54,  2.92s/it]
 71%|███████   | 1824/2562 [1:33:09<35:33,  2.89s/it]
 71%|███████   | 1825/2562 [1:33:12<37:02,  3.02s/it]
 71%|███████▏  | 1826/2562 [1:33:15<36:40,  2.99s/it]
 71%|███████▏  | 1827/2562 [1:33:18<36:12,  2.96s/it]
 71%|███████▏  | 1828/2562 [1:33:21<36:26,  2.98s/it]
 71%|███████▏  | 1829/2562 [1:33:24<35:36,  2.91s/it]
 71%|███████▏  | 1830/2562 [1:33:27<35:20,  2.90s/it]
                                                     
{'loss': 0.8311, 'grad_norm': 5.489665508270264, 'learning_rate': 4.144106436800453e-06, 'epoch': 2.14}

 71%|███████▏  | 1830/2562 [1:33:27<35:20,  2.90s/it]
 71%|███████▏  | 1831/2562 [1:33:30<34:34,  2.84s/it]
 72%|███████▏  | 1832/2562 [1:33:32<34:15,  2.82s/it]
 72%|███████▏  | 1833/2562 [1:33:35<34:50,  2.87s/it]
 72%|███████▏  | 1834/2562 [1:33:38<34:28,  2.84s/it]
 72%|███████▏  | 1835/2562 [1:33:41<34:14,  2.83s/it]
 72%|███████▏  | 1836/2562 [1:33:44<35:06,  2.90s/it]
 72%|███████▏  | 1837/2562 [1:33:47<34:17,  2.84s/it]
 72%|███████▏  | 1838/2562 [1:33:49<33:57,  2.81s/it]
 72%|███████▏  | 1839/2562 [1:33:52<34:31,  2.86s/it]
 72%|███████▏  | 1840/2562 [1:33:55<34:30,  2.87s/it]
                                                     
{'loss': 0.8185, 'grad_norm': 4.973864555358887, 'learning_rate': 4.039928396563983e-06, 'epoch': 2.15}

 72%|███████▏  | 1840/2562 [1:33:55<34:30,  2.87s/it]
 72%|███████▏  | 1841/2562 [1:33:58<34:17,  2.85s/it]
 72%|███████▏  | 1842/2562 [1:34:01<34:18,  2.86s/it]
 72%|███████▏  | 1843/2562 [1:34:04<35:19,  2.95s/it]
 72%|███████▏  | 1844/2562 [1:34:07<35:29,  2.97s/it]
 72%|███████▏  | 1845/2562 [1:34:10<35:36,  2.98s/it]
 72%|███████▏  | 1846/2562 [1:34:13<34:42,  2.91s/it]
 72%|███████▏  | 1847/2562 [1:34:16<35:05,  2.95s/it]
 72%|███████▏  | 1848/2562 [1:34:19<35:17,  2.97s/it]
 72%|███████▏  | 1849/2562 [1:34:22<35:06,  2.96s/it]
 72%|███████▏  | 1850/2562 [1:34:25<35:04,  2.96s/it]
                                                     
{'loss': 0.8145, 'grad_norm': 4.900149345397949, 'learning_rate': 3.936744069325797e-06, 'epoch': 2.17}

 72%|███████▏  | 1850/2562 [1:34:25<35:04,  2.96s/it]
 72%|███████▏  | 1851/2562 [1:34:28<34:29,  2.91s/it]
 72%|███████▏  | 1852/2562 [1:34:30<33:58,  2.87s/it]
 72%|███████▏  | 1853/2562 [1:34:33<33:28,  2.83s/it]
 72%|███████▏  | 1854/2562 [1:34:36<34:36,  2.93s/it]
 72%|███████▏  | 1855/2562 [1:34:39<33:51,  2.87s/it]
 72%|███████▏  | 1856/2562 [1:34:42<34:34,  2.94s/it]
 72%|███████▏  | 1857/2562 [1:34:45<33:47,  2.88s/it]
 73%|███████▎  | 1858/2562 [1:34:48<33:33,  2.86s/it]
 73%|███████▎  | 1859/2562 [1:34:51<33:38,  2.87s/it]
 73%|███████▎  | 1860/2562 [1:34:53<33:07,  2.83s/it]
                                                     
{'loss': 0.8229, 'grad_norm': 4.784515380859375, 'learning_rate': 3.834570658840152e-06, 'epoch': 2.18}

 73%|███████▎  | 1860/2562 [1:34:53<33:07,  2.83s/it]
 73%|███████▎  | 1861/2562 [1:34:56<33:13,  2.84s/it]
 73%|███████▎  | 1862/2562 [1:34:59<33:30,  2.87s/it]
 73%|███████▎  | 1863/2562 [1:35:02<33:46,  2.90s/it]
 73%|███████▎  | 1864/2562 [1:35:05<33:38,  2.89s/it]
 73%|███████▎  | 1865/2562 [1:35:08<34:01,  2.93s/it]
 73%|███████▎  | 1866/2562 [1:35:11<33:10,  2.86s/it]
 73%|███████▎  | 1867/2562 [1:35:14<33:20,  2.88s/it]
 73%|███████▎  | 1868/2562 [1:35:16<33:08,  2.87s/it]
 73%|███████▎  | 1869/2562 [1:35:19<32:59,  2.86s/it]
 73%|███████▎  | 1870/2562 [1:35:22<32:44,  2.84s/it]
                                                     
{'loss': 0.7878, 'grad_norm': 4.94583797454834, 'learning_rate': 3.733425200312797e-06, 'epoch': 2.19}

 73%|███████▎  | 1870/2562 [1:35:22<32:44,  2.84s/it]
 73%|███████▎  | 1871/2562 [1:35:25<33:28,  2.91s/it]
 73%|███████▎  | 1872/2562 [1:35:28<33:21,  2.90s/it]
 73%|███████▎  | 1873/2562 [1:35:31<32:59,  2.87s/it]
 73%|███████▎  | 1874/2562 [1:35:34<33:34,  2.93s/it]
 73%|███████▎  | 1875/2562 [1:35:37<32:55,  2.88s/it]
 73%|███████▎  | 1876/2562 [1:35:40<33:11,  2.90s/it]
 73%|███████▎  | 1877/2562 [1:35:42<33:10,  2.91s/it]
 73%|███████▎  | 1878/2562 [1:35:45<32:57,  2.89s/it]
 73%|███████▎  | 1879/2562 [1:35:48<32:28,  2.85s/it]
 73%|███████▎  | 1880/2562 [1:35:51<32:56,  2.90s/it]
                                                     
{'loss': 0.7933, 'grad_norm': 4.659102439880371, 'learning_rate': 3.633324557560747e-06, 'epoch': 2.2}

 73%|███████▎  | 1880/2562 [1:35:51<32:56,  2.90s/it]
 73%|███████▎  | 1881/2562 [1:35:54<33:26,  2.95s/it]
 73%|███████▎  | 1882/2562 [1:35:57<32:32,  2.87s/it]
 73%|███████▎  | 1883/2562 [1:36:00<33:10,  2.93s/it]
 74%|███████▎  | 1884/2562 [1:36:03<33:18,  2.95s/it]
 74%|███████▎  | 1885/2562 [1:36:06<32:55,  2.92s/it]
 74%|███████▎  | 1886/2562 [1:36:09<33:05,  2.94s/it]
 74%|███████▎  | 1887/2562 [1:36:12<32:41,  2.91s/it]
 74%|███████▎  | 1888/2562 [1:36:14<32:15,  2.87s/it]
 74%|███████▎  | 1889/2562 [1:36:17<32:24,  2.89s/it]
 74%|███████▍  | 1890/2562 [1:36:20<31:59,  2.86s/it]
                                                     
{'loss': 0.8034, 'grad_norm': 4.737246036529541, 'learning_rate': 3.5342854202005696e-06, 'epoch': 2.21}

 74%|███████▍  | 1890/2562 [1:36:20<31:59,  2.86s/it]
 74%|███████▍  | 1891/2562 [1:36:23<31:34,  2.82s/it]
 74%|███████▍  | 1892/2562 [1:36:26<31:54,  2.86s/it]
 74%|███████▍  | 1893/2562 [1:36:29<31:51,  2.86s/it]
 74%|███████▍  | 1894/2562 [1:36:31<31:43,  2.85s/it]
 74%|███████▍  | 1895/2562 [1:36:34<32:17,  2.90s/it]
 74%|███████▍  | 1896/2562 [1:36:37<31:52,  2.87s/it]
 74%|███████▍  | 1897/2562 [1:36:40<31:48,  2.87s/it]
 74%|███████▍  | 1898/2562 [1:36:43<32:03,  2.90s/it]
 74%|███████▍  | 1899/2562 [1:36:46<31:32,  2.86s/it]
 74%|███████▍  | 1900/2562 [1:36:49<32:13,  2.92s/it]
                                                     
{'loss': 0.8193, 'grad_norm': 4.860280990600586, 'learning_rate': 3.4363243008657842e-06, 'epoch': 2.22}

 74%|███████▍  | 1900/2562 [1:36:49<32:13,  2.92s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.67it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.73it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.22it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.89it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.56it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.72it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.76it/s]

 10%|█         | 14/140 [00:01<00:12,  9.76it/s]

 11%|█         | 15/140 [00:01<00:12,  9.80it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.81it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.83it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.61it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.52it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.48it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.44it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.55it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.44it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.13it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.19it/s]

 20%|██        | 28/140 [00:02<00:11,  9.51it/s]

 21%|██        | 29/140 [00:02<00:11,  9.37it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.06it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.06it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.25it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.70it/s]

 25%|██▌       | 35/140 [00:03<00:10,  9.77it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.77it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.81it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.84it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.95it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.94it/s]

 31%|███       | 43/140 [00:04<00:09, 10.01it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.97it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.91it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.67it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.33it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.30it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.19it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.26it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.44it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.44it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.43it/s]

 40%|████      | 56/140 [00:05<00:08,  9.72it/s]

 41%|████      | 57/140 [00:05<00:08,  9.51it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.42it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.63it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.83it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.64it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.57it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.52it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.50it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.72it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.82it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.83it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.87it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.88it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.88it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.85it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.61it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.22it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.37it/s]

 57%|█████▋    | 80/140 [00:08<00:06,  9.55it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.20it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.15it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.04it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.05it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.28it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.16it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.17it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.05it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.99it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.07it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.27it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.45it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.27it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.04it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.13it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.30it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.34it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.50it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.30it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.37it/s]

 74%|███████▍  | 104/140 [00:10<00:03,  9.70it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.73it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.61it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.50it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.59it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.66it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.85it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.85it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.54it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.44it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.54it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.31it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.09it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.08it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.21it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.37it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.26it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.04it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.82it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.58it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.69it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  8.96it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.51it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.67it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.50it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.53it/s]

100%|██████████| 140/140 [00:14<00:00,  8.14it/s]
                                                     

                                                 
{'eval_loss': 0.9166725277900696, 'eval_runtime': 15.1232, 'eval_samples_per_second': 18.515, 'eval_steps_per_second': 9.257, 'epoch': 2.22}

 74%|███████▍  | 1900/2562 [1:37:04<32:13,  2.92s/it]

100%|██████████| 140/140 [00:14<00:00,  8.14it/s]

                                                 
 74%|███████▍  | 1901/2562 [1:37:07<1:21:54,  7.43s/it]
 74%|███████▍  | 1902/2562 [1:37:10<1:06:28,  6.04s/it]
 74%|███████▍  | 1903/2562 [1:37:13<56:07,  5.11s/it]  
 74%|███████▍  | 1904/2562 [1:37:15<48:29,  4.42s/it]
 74%|███████▍  | 1905/2562 [1:37:18<43:14,  3.95s/it]
 74%|███████▍  | 1906/2562 [1:37:21<40:32,  3.71s/it]
 74%|███████▍  | 1907/2562 [1:37:24<38:08,  3.49s/it]
 74%|███████▍  | 1908/2562 [1:37:27<36:13,  3.32s/it]
 75%|███████▍  | 1909/2562 [1:37:30<35:13,  3.24s/it]
 75%|███████▍  | 1910/2562 [1:37:33<34:01,  3.13s/it]
                                                     
{'loss': 0.799, 'grad_norm': 4.866443157196045, 'learning_rate': 3.3394575324537327e-06, 'epoch': 2.24}

 75%|███████▍  | 1910/2562 [1:37:33<34:01,  3.13s/it]
 75%|███████▍  | 1911/2562 [1:37:36<33:32,  3.09s/it]
 75%|███████▍  | 1912/2562 [1:37:39<32:58,  3.04s/it]
 75%|███████▍  | 1913/2562 [1:37:42<32:58,  3.05s/it]
 75%|███████▍  | 1914/2562 [1:37:45<32:00,  2.96s/it]
 75%|███████▍  | 1915/2562 [1:37:48<31:54,  2.96s/it]
 75%|███████▍  | 1916/2562 [1:37:51<31:17,  2.91s/it]
 75%|███████▍  | 1917/2562 [1:37:54<30:49,  2.87s/it]
 75%|███████▍  | 1918/2562 [1:37:56<30:57,  2.88s/it]
 75%|███████▍  | 1919/2562 [1:38:00<31:25,  2.93s/it]
 75%|███████▍  | 1920/2562 [1:38:02<30:45,  2.87s/it]
                                                     
{'loss': 0.8017, 'grad_norm': 4.993982315063477, 'learning_rate': 3.2437012654024057e-06, 'epoch': 2.25}

 75%|███████▍  | 1920/2562 [1:38:02<30:45,  2.87s/it]
 75%|███████▍  | 1921/2562 [1:38:05<31:02,  2.91s/it]
 75%|███████▌  | 1922/2562 [1:38:08<30:43,  2.88s/it]
 75%|███████▌  | 1923/2562 [1:38:11<30:27,  2.86s/it]
 75%|███████▌  | 1924/2562 [1:38:14<30:58,  2.91s/it]
 75%|███████▌  | 1925/2562 [1:38:17<30:37,  2.88s/it]
 75%|███████▌  | 1926/2562 [1:38:20<31:34,  2.98s/it]
 75%|███████▌  | 1927/2562 [1:38:23<31:19,  2.96s/it]
 75%|███████▌  | 1928/2562 [1:38:26<30:58,  2.93s/it]
 75%|███████▌  | 1929/2562 [1:38:29<31:13,  2.96s/it]
 75%|███████▌  | 1930/2562 [1:38:32<30:54,  2.93s/it]
                                                     
{'loss': 0.7921, 'grad_norm': 5.113594055175781, 'learning_rate': 3.1490714649977196e-06, 'epoch': 2.26}

 75%|███████▌  | 1930/2562 [1:38:32<30:54,  2.93s/it]
 75%|███████▌  | 1931/2562 [1:38:34<30:32,  2.90s/it]
 75%|███████▌  | 1932/2562 [1:38:37<30:21,  2.89s/it]
 75%|███████▌  | 1933/2562 [1:38:40<30:42,  2.93s/it]
 75%|███████▌  | 1934/2562 [1:38:43<30:24,  2.90s/it]
 76%|███████▌  | 1935/2562 [1:38:46<30:45,  2.94s/it]
 76%|███████▌  | 1936/2562 [1:38:49<30:31,  2.93s/it]
 76%|███████▌  | 1937/2562 [1:38:52<29:43,  2.85s/it]
 76%|███████▌  | 1938/2562 [1:38:55<29:54,  2.88s/it]
 76%|███████▌  | 1939/2562 [1:38:57<29:34,  2.85s/it]
 76%|███████▌  | 1940/2562 [1:39:00<29:22,  2.83s/it]
                                                     
{'loss': 0.8087, 'grad_norm': 4.95025110244751, 'learning_rate': 3.0555839087116547e-06, 'epoch': 2.27}

 76%|███████▌  | 1940/2562 [1:39:00<29:22,  2.83s/it]
 76%|███████▌  | 1941/2562 [1:39:03<29:49,  2.88s/it]
 76%|███████▌  | 1942/2562 [1:39:06<29:33,  2.86s/it]
 76%|███████▌  | 1943/2562 [1:39:09<29:33,  2.87s/it]
 76%|███████▌  | 1944/2562 [1:39:12<29:57,  2.91s/it]
 76%|███████▌  | 1945/2562 [1:39:15<30:39,  2.98s/it]
 76%|███████▌  | 1946/2562 [1:39:18<30:33,  2.98s/it]
 76%|███████▌  | 1947/2562 [1:39:21<30:13,  2.95s/it]
 76%|███████▌  | 1948/2562 [1:39:24<29:51,  2.92s/it]
 76%|███████▌  | 1949/2562 [1:39:27<29:45,  2.91s/it]
 76%|███████▌  | 1950/2562 [1:39:30<30:04,  2.95s/it]
                                                     
{'loss': 0.8022, 'grad_norm': 4.9889140129089355, 'learning_rate': 2.963254183571682e-06, 'epoch': 2.28}

 76%|███████▌  | 1950/2562 [1:39:30<30:04,  2.95s/it]
 76%|███████▌  | 1951/2562 [1:39:33<29:52,  2.93s/it]
 76%|███████▌  | 1952/2562 [1:39:36<30:00,  2.95s/it]
 76%|███████▌  | 1953/2562 [1:39:39<30:10,  2.97s/it]
 76%|███████▋  | 1954/2562 [1:39:41<29:39,  2.93s/it]
 76%|███████▋  | 1955/2562 [1:39:44<29:29,  2.92s/it]
 76%|███████▋  | 1956/2562 [1:39:47<29:16,  2.90s/it]
 76%|███████▋  | 1957/2562 [1:39:50<28:57,  2.87s/it]
 76%|███████▋  | 1958/2562 [1:39:53<29:05,  2.89s/it]
 76%|███████▋  | 1959/2562 [1:39:56<29:34,  2.94s/it]
 77%|███████▋  | 1960/2562 [1:39:59<29:17,  2.92s/it]
                                                     
{'loss': 0.7812, 'grad_norm': 4.794127464294434, 'learning_rate': 2.872097683561986e-06, 'epoch': 2.29}

 77%|███████▋  | 1960/2562 [1:39:59<29:17,  2.92s/it]
 77%|███████▋  | 1961/2562 [1:40:02<28:51,  2.88s/it]
 77%|███████▋  | 1962/2562 [1:40:05<29:06,  2.91s/it]
 77%|███████▋  | 1963/2562 [1:40:07<28:49,  2.89s/it]
 77%|███████▋  | 1964/2562 [1:40:10<28:31,  2.86s/it]
 77%|███████▋  | 1965/2562 [1:40:13<28:48,  2.90s/it]
 77%|███████▋  | 1966/2562 [1:40:16<28:47,  2.90s/it]
 77%|███████▋  | 1967/2562 [1:40:19<28:52,  2.91s/it]
 77%|███████▋  | 1968/2562 [1:40:22<28:59,  2.93s/it]
 77%|███████▋  | 1969/2562 [1:40:25<28:30,  2.89s/it]
 77%|███████▋  | 1970/2562 [1:40:28<28:09,  2.85s/it]
                                                     
{'loss': 0.8087, 'grad_norm': 5.115259647369385, 'learning_rate': 2.782129607056848e-06, 'epoch': 2.31}

 77%|███████▋  | 1970/2562 [1:40:28<28:09,  2.85s/it]
 77%|███████▋  | 1971/2562 [1:40:31<28:20,  2.88s/it]
 77%|███████▋  | 1972/2562 [1:40:33<28:11,  2.87s/it]
 77%|███████▋  | 1973/2562 [1:40:36<28:07,  2.86s/it]
 77%|███████▋  | 1974/2562 [1:40:39<28:31,  2.91s/it]
 77%|███████▋  | 1975/2562 [1:40:42<28:34,  2.92s/it]
 77%|███████▋  | 1976/2562 [1:40:45<28:26,  2.91s/it]
 77%|███████▋  | 1977/2562 [1:40:48<28:08,  2.89s/it]
 77%|███████▋  | 1978/2562 [1:40:51<27:40,  2.84s/it]
 77%|███████▋  | 1979/2562 [1:40:54<27:58,  2.88s/it]
 77%|███████▋  | 1980/2562 [1:40:57<27:57,  2.88s/it]
                                                     
{'loss': 0.7926, 'grad_norm': 5.039003849029541, 'learning_rate': 2.6933649542866326e-06, 'epoch': 2.32}

 77%|███████▋  | 1980/2562 [1:40:57<27:57,  2.88s/it]
 77%|███████▋  | 1981/2562 [1:40:59<27:22,  2.83s/it]
 77%|███████▋  | 1982/2562 [1:41:02<27:47,  2.87s/it]
 77%|███████▋  | 1983/2562 [1:41:05<27:26,  2.84s/it]
 77%|███████▋  | 1984/2562 [1:41:08<27:27,  2.85s/it]
 77%|███████▋  | 1985/2562 [1:41:11<27:35,  2.87s/it]
 78%|███████▊  | 1986/2562 [1:41:14<27:43,  2.89s/it]
 78%|███████▊  | 1987/2562 [1:41:17<27:27,  2.86s/it]
 78%|███████▊  | 1988/2562 [1:41:20<27:49,  2.91s/it]
 78%|███████▊  | 1989/2562 [1:41:22<27:38,  2.89s/it]
 78%|███████▊  | 1990/2562 [1:41:25<27:21,  2.87s/it]
                                                     
{'loss': 0.8045, 'grad_norm': 4.693707466125488, 'learning_rate': 2.6058185248368317e-06, 'epoch': 2.33}

 78%|███████▊  | 1990/2562 [1:41:25<27:21,  2.87s/it]
 78%|███████▊  | 1991/2562 [1:41:28<27:25,  2.88s/it]
 78%|███████▊  | 1992/2562 [1:41:31<26:49,  2.82s/it]
 78%|███████▊  | 1993/2562 [1:41:34<26:44,  2.82s/it]
 78%|███████▊  | 1994/2562 [1:41:36<26:45,  2.83s/it]
 78%|███████▊  | 1995/2562 [1:41:39<26:55,  2.85s/it]
 78%|███████▊  | 1996/2562 [1:41:42<26:50,  2.85s/it]
 78%|███████▊  | 1997/2562 [1:41:45<27:23,  2.91s/it]
 78%|███████▊  | 1998/2562 [1:41:48<27:02,  2.88s/it]
 78%|███████▊  | 1999/2562 [1:41:51<27:03,  2.88s/it]
 78%|███████▊  | 2000/2562 [1:41:54<26:51,  2.87s/it]
                                                     
{'loss': 0.7746, 'grad_norm': 5.121265888214111, 'learning_rate': 2.519504915180555e-06, 'epoch': 2.34}

 78%|███████▊  | 2000/2562 [1:41:54<26:51,  2.87s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.59it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.74it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.22it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.89it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.56it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.74it/s]

 10%|█         | 14/140 [00:01<00:12,  9.79it/s]

 11%|█         | 15/140 [00:01<00:12,  9.83it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.82it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.85it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.64it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.59it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.68it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.43it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.38it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.52it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.37it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.06it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.15it/s]

 20%|██        | 28/140 [00:02<00:11,  9.53it/s]

 21%|██        | 29/140 [00:02<00:11,  9.39it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.08it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.08it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.31it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.76it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.83it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.84it/s]

 28%|██▊       | 39/140 [00:03<00:10,  9.93it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.92it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 30%|███       | 42/140 [00:04<00:09,  9.91it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.94it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.91it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.68it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.34it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.31it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.21it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.26it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.45it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.44it/s]

 40%|████      | 56/140 [00:05<00:08,  9.70it/s]

 41%|████      | 57/140 [00:05<00:08,  9.50it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.39it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.60it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.82it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.65it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.59it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.56it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.54it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.74it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.77it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.78it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.75it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.81it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.84it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.83it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.85it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.59it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.19it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.33it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.29it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.21it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.12it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.15it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.36it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.24it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.22it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.13it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.11it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.00it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.09it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.30it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.44it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.28it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.05it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.16it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.30it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.35it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.49it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.30it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.35it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.49it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.75it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.62it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.52it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.60it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.66it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.84it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.85it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.54it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.45it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.58it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.35it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.33it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.44it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.15it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.12it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.29it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.24it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.39it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.28it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.07it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.83it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.55it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.68it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.02it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.49it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.65it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.51it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.52it/s]

100%|██████████| 140/140 [00:14<00:00,  8.14it/s]
                                                     

                                                 
{'eval_loss': 0.9115705490112305, 'eval_runtime': 15.1093, 'eval_samples_per_second': 18.532, 'eval_steps_per_second': 9.266, 'epoch': 2.34}

 78%|███████▊  | 2000/2562 [1:42:09<26:51,  2.87s/it]

100%|██████████| 140/140 [00:14<00:00,  8.14it/s]

                                                 /opt/conda/lib/python3.11/site-packages/peft/utils/save_and_load.py:372: UserWarning: Could not find a config file in  - will assume that the vocabulary was not modified.
  warnings.warn(

 78%|███████▊  | 2001/2562 [1:42:13<1:12:46,  7.78s/it]
 78%|███████▊  | 2002/2562 [1:42:16<59:32,  6.38s/it]  
 78%|███████▊  | 2003/2562 [1:42:19<49:42,  5.34s/it]
 78%|███████▊  | 2004/2562 [1:42:22<42:25,  4.56s/it]
 78%|███████▊  | 2005/2562 [1:42:25<37:56,  4.09s/it]
 78%|███████▊  | 2006/2562 [1:42:28<34:07,  3.68s/it]
 78%|███████▊  | 2007/2562 [1:42:30<31:41,  3.43s/it]
 78%|███████▊  | 2008/2562 [1:42:33<30:33,  3.31s/it]
 78%|███████▊  | 2009/2562 [1:42:36<29:04,  3.15s/it]
 78%|███████▊  | 2010/2562 [1:42:39<28:30,  3.10s/it]
                                                     
{'loss': 0.8155, 'grad_norm': 4.672001838684082, 'learning_rate': 2.4344385162448924e-06, 'epoch': 2.35}

 78%|███████▊  | 2010/2562 [1:42:39<28:30,  3.10s/it]
 78%|███████▊  | 2011/2562 [1:42:42<28:04,  3.06s/it]
 79%|███████▊  | 2012/2562 [1:42:45<27:24,  2.99s/it]
 79%|███████▊  | 2013/2562 [1:42:48<26:52,  2.94s/it]
 79%|███████▊  | 2014/2562 [1:42:51<27:07,  2.97s/it]
 79%|███████▊  | 2015/2562 [1:42:54<26:33,  2.91s/it]
 79%|███████▊  | 2016/2562 [1:42:56<26:08,  2.87s/it]
 79%|███████▊  | 2017/2562 [1:42:59<26:33,  2.92s/it]
 79%|███████▉  | 2018/2562 [1:43:02<26:25,  2.92s/it]
 79%|███████▉  | 2019/2562 [1:43:05<26:00,  2.87s/it]
 79%|███████▉  | 2020/2562 [1:43:08<26:20,  2.92s/it]
                                                     
{'loss': 0.7971, 'grad_norm': 5.082230567932129, 'learning_rate': 2.350633511011511e-06, 'epoch': 2.37}

 79%|███████▉  | 2020/2562 [1:43:08<26:20,  2.92s/it]
 79%|███████▉  | 2021/2562 [1:43:11<25:48,  2.86s/it]
 79%|███████▉  | 2022/2562 [1:43:14<25:38,  2.85s/it]
 79%|███████▉  | 2023/2562 [1:43:17<25:54,  2.88s/it]
 79%|███████▉  | 2024/2562 [1:43:19<25:28,  2.84s/it]
 79%|███████▉  | 2025/2562 [1:43:22<25:27,  2.84s/it]
 79%|███████▉  | 2026/2562 [1:43:25<26:08,  2.93s/it]
 79%|███████▉  | 2027/2562 [1:43:28<25:41,  2.88s/it]
 79%|███████▉  | 2028/2562 [1:43:31<25:45,  2.89s/it]
 79%|███████▉  | 2029/2562 [1:43:34<26:20,  2.96s/it]
 79%|███████▉  | 2030/2562 [1:43:37<25:48,  2.91s/it]
                                                     
{'loss': 0.7868, 'grad_norm': 4.738226413726807, 'learning_rate': 2.2681038721519768e-06, 'epoch': 2.38}

 79%|███████▉  | 2030/2562 [1:43:37<25:48,  2.91s/it]
 79%|███████▉  | 2031/2562 [1:43:40<25:49,  2.92s/it]
 79%|███████▉  | 2032/2562 [1:43:43<25:38,  2.90s/it]
 79%|███████▉  | 2033/2562 [1:43:46<25:32,  2.90s/it]
 79%|███████▉  | 2034/2562 [1:43:49<25:43,  2.92s/it]
 79%|███████▉  | 2035/2562 [1:43:51<25:20,  2.88s/it]
 79%|███████▉  | 2036/2562 [1:43:54<25:15,  2.88s/it]
 80%|███████▉  | 2037/2562 [1:43:57<25:21,  2.90s/it]
 80%|███████▉  | 2038/2562 [1:44:00<25:05,  2.87s/it]
 80%|███████▉  | 2039/2562 [1:44:03<24:38,  2.83s/it]
 80%|███████▉  | 2040/2562 [1:44:06<25:23,  2.92s/it]
                                                     
{'loss': 0.8061, 'grad_norm': 5.041847229003906, 'learning_rate': 2.186863359698108e-06, 'epoch': 2.39}

 80%|███████▉  | 2040/2562 [1:44:06<25:23,  2.92s/it]
 80%|███████▉  | 2041/2562 [1:44:09<24:58,  2.88s/it]
 80%|███████▉  | 2042/2562 [1:44:12<25:15,  2.91s/it]
 80%|███████▉  | 2043/2562 [1:44:15<26:01,  3.01s/it]
 80%|███████▉  | 2044/2562 [1:44:18<25:41,  2.98s/it]
 80%|███████▉  | 2045/2562 [1:44:21<25:25,  2.95s/it]
 80%|███████▉  | 2046/2562 [1:44:24<25:20,  2.95s/it]
 80%|███████▉  | 2047/2562 [1:44:27<25:11,  2.94s/it]
 80%|███████▉  | 2048/2562 [1:44:29<24:45,  2.89s/it]
 80%|███████▉  | 2049/2562 [1:44:32<24:42,  2.89s/it]
 80%|████████  | 2050/2562 [1:44:35<24:41,  2.89s/it]
                                                     
{'loss': 0.7816, 'grad_norm': 5.128898620605469, 'learning_rate': 2.106925518747779e-06, 'epoch': 2.4}

 80%|████████  | 2050/2562 [1:44:35<24:41,  2.89s/it]
 80%|████████  | 2051/2562 [1:44:38<24:42,  2.90s/it]
 80%|████████  | 2052/2562 [1:44:41<25:03,  2.95s/it]
 80%|████████  | 2053/2562 [1:44:44<24:54,  2.94s/it]
 80%|████████  | 2054/2562 [1:44:47<24:35,  2.90s/it]
 80%|████████  | 2055/2562 [1:44:50<24:44,  2.93s/it]
 80%|████████  | 2056/2562 [1:44:53<24:11,  2.87s/it]
 80%|████████  | 2057/2562 [1:44:55<24:05,  2.86s/it]
 80%|████████  | 2058/2562 [1:44:58<24:25,  2.91s/it]
 80%|████████  | 2059/2562 [1:45:01<24:10,  2.88s/it]
 80%|████████  | 2060/2562 [1:45:04<24:00,  2.87s/it]
                                                     
{'loss': 0.7798, 'grad_norm': 4.893022060394287, 'learning_rate': 2.0283036772065712e-06, 'epoch': 2.41}

 80%|████████  | 2060/2562 [1:45:04<24:00,  2.87s/it]
 80%|████████  | 2061/2562 [1:45:07<24:22,  2.92s/it]
 80%|████████  | 2062/2562 [1:45:10<24:32,  2.94s/it]
 81%|████████  | 2063/2562 [1:45:13<24:27,  2.94s/it]
 81%|████████  | 2064/2562 [1:45:16<24:28,  2.95s/it]
 81%|████████  | 2065/2562 [1:45:19<24:19,  2.94s/it]
 81%|████████  | 2066/2562 [1:45:22<24:08,  2.92s/it]
 81%|████████  | 2067/2562 [1:45:25<24:29,  2.97s/it]
 81%|████████  | 2068/2562 [1:45:28<24:09,  2.93s/it]
 81%|████████  | 2069/2562 [1:45:31<24:12,  2.95s/it]
 81%|████████  | 2070/2562 [1:45:34<24:07,  2.94s/it]
                                                     
{'loss': 0.7675, 'grad_norm': 4.870106220245361, 'learning_rate': 1.9510109435656457e-06, 'epoch': 2.42}

 81%|████████  | 2070/2562 [1:45:34<24:07,  2.94s/it]
 81%|████████  | 2071/2562 [1:45:36<23:45,  2.90s/it]
 81%|████████  | 2072/2562 [1:45:39<23:28,  2.87s/it]
 81%|████████  | 2073/2562 [1:45:42<23:41,  2.91s/it]
 81%|████████  | 2074/2562 [1:45:45<23:19,  2.87s/it]
 81%|████████  | 2075/2562 [1:45:48<23:03,  2.84s/it]
 81%|████████  | 2076/2562 [1:45:51<23:20,  2.88s/it]
 81%|████████  | 2077/2562 [1:45:54<23:22,  2.89s/it]
 81%|████████  | 2078/2562 [1:45:57<23:09,  2.87s/it]
 81%|████████  | 2079/2562 [1:46:00<23:33,  2.93s/it]
 81%|████████  | 2080/2562 [1:46:02<23:11,  2.89s/it]
                                                     
{'loss': 0.7844, 'grad_norm': 4.866476058959961, 'learning_rate': 1.8750602047161603e-06, 'epoch': 2.44}

 81%|████████  | 2080/2562 [1:46:02<23:11,  2.89s/it]
 81%|████████  | 2081/2562 [1:46:05<23:20,  2.91s/it]
 81%|████████▏ | 2082/2562 [1:46:08<23:44,  2.97s/it]
 81%|████████▏ | 2083/2562 [1:46:11<23:51,  2.99s/it]
 81%|████████▏ | 2084/2562 [1:46:14<23:33,  2.96s/it]
 81%|████████▏ | 2085/2562 [1:46:17<23:37,  2.97s/it]
 81%|████████▏ | 2086/2562 [1:46:20<23:08,  2.92s/it]
 81%|████████▏ | 2087/2562 [1:46:23<22:46,  2.88s/it]
 81%|████████▏ | 2088/2562 [1:46:26<22:49,  2.89s/it]
 82%|████████▏ | 2089/2562 [1:46:29<22:50,  2.90s/it]
 82%|████████▏ | 2090/2562 [1:46:32<22:47,  2.90s/it]
                                                     
{'loss': 0.7814, 'grad_norm': 4.954866886138916, 'learning_rate': 1.8004641238006815e-06, 'epoch': 2.45}

 82%|████████▏ | 2090/2562 [1:46:32<22:47,  2.90s/it]
 82%|████████▏ | 2091/2562 [1:46:35<23:21,  2.98s/it]
 82%|████████▏ | 2092/2562 [1:46:38<22:59,  2.94s/it]
 82%|████████▏ | 2093/2562 [1:46:41<23:03,  2.95s/it]
 82%|████████▏ | 2094/2562 [1:46:44<23:31,  3.02s/it]
 82%|████████▏ | 2095/2562 [1:46:47<23:01,  2.96s/it]
 82%|████████▏ | 2096/2562 [1:46:49<22:39,  2.92s/it]
 82%|████████▏ | 2097/2562 [1:46:52<22:46,  2.94s/it]
 82%|████████▏ | 2098/2562 [1:46:55<22:39,  2.93s/it]
 82%|████████▏ | 2099/2562 [1:46:58<22:34,  2.93s/it]
 82%|████████▏ | 2100/2562 [1:47:01<22:27,  2.92s/it]
                                                     
{'loss': 0.7889, 'grad_norm': 4.619852066040039, 'learning_rate': 1.7272351381018792e-06, 'epoch': 2.46}

 82%|████████▏ | 2100/2562 [1:47:01<22:27,  2.92s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.81it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.73it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.20it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.87it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.53it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.69it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.71it/s]

 10%|█         | 14/140 [00:01<00:12,  9.75it/s]

 11%|█         | 15/140 [00:01<00:12,  9.76it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.77it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.79it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.56it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.48it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.46it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.42it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.53it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.38it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.09it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.16it/s]

 20%|██        | 28/140 [00:02<00:11,  9.47it/s]

 21%|██        | 29/140 [00:02<00:11,  9.32it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.03it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.03it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.24it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.69it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.74it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.77it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.81it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.88it/s]

 29%|██▉       | 41/140 [00:04<00:10,  9.88it/s]

 30%|███       | 42/140 [00:04<00:09,  9.90it/s]

 31%|███       | 43/140 [00:04<00:09,  9.91it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.87it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.84it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.59it/s]

 34%|███▎      | 47/140 [00:04<00:10,  9.22it/s]

 34%|███▍      | 48/140 [00:04<00:10,  9.19it/s]

 35%|███▌      | 49/140 [00:05<00:10,  9.06it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.14it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.35it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.41it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.43it/s]

 40%|████      | 56/140 [00:05<00:08,  9.70it/s]

 41%|████      | 57/140 [00:05<00:08,  9.49it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.38it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.49it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.60it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.80it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.61it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.53it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.45it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.41it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.65it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.73it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.75it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.79it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.82it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.84it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.82it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.83it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.59it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.21it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.35it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.28it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.20it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.11it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.12it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.34it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.20it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.18it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.09it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.11it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.01it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.10it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.31it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.43it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.27it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.01it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.10it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.27it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.32it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.46it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.28it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.33it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.50it/s]

 75%|███████▌  | 105/140 [00:11<00:03,  9.72it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.59it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.50it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.58it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.65it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.68it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.83it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.84it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.53it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.56it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.33it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.30it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.12it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.10it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.26it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.22it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.37it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.27it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.05it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.81it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.55it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.68it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  8.98it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.45it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.62it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.48it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.47it/s]

100%|██████████| 140/140 [00:14<00:00,  8.15it/s]
                                                     

                                                 
{'eval_loss': 0.9049318432807922, 'eval_runtime': 15.1527, 'eval_samples_per_second': 18.479, 'eval_steps_per_second': 9.239, 'epoch': 2.46}

 82%|████████▏ | 2100/2562 [1:47:16<22:27,  2.92s/it]

100%|██████████| 140/140 [00:14<00:00,  8.15it/s]

                                                 
 82%|████████▏ | 2101/2562 [1:47:19<57:16,  7.45s/it]
 82%|████████▏ | 2102/2562 [1:47:22<46:58,  6.13s/it]
 82%|████████▏ | 2103/2562 [1:47:25<39:58,  5.23s/it]
 82%|████████▏ | 2104/2562 [1:47:28<34:29,  4.52s/it]
 82%|████████▏ | 2105/2562 [1:47:31<30:39,  4.03s/it]
 82%|████████▏ | 2106/2562 [1:47:34<28:39,  3.77s/it]
 82%|████████▏ | 2107/2562 [1:47:37<26:17,  3.47s/it]
 82%|████████▏ | 2108/2562 [1:47:40<25:08,  3.32s/it]
 82%|████████▏ | 2109/2562 [1:47:43<24:15,  3.21s/it]
 82%|████████▏ | 2110/2562 [1:47:46<23:08,  3.07s/it]
                                                     
{'loss': 0.806, 'grad_norm': 5.177557945251465, 'learning_rate': 1.6553854569688632e-06, 'epoch': 2.47}

 82%|████████▏ | 2110/2562 [1:47:46<23:08,  3.07s/it]
 82%|████████▏ | 2111/2562 [1:47:49<23:04,  3.07s/it]
 82%|████████▏ | 2112/2562 [1:47:52<22:49,  3.04s/it]
 82%|████████▏ | 2113/2562 [1:47:55<22:23,  2.99s/it]
 83%|████████▎ | 2114/2562 [1:47:58<22:29,  3.01s/it]
 83%|████████▎ | 2115/2562 [1:48:01<22:11,  2.98s/it]
 83%|████████▎ | 2116/2562 [1:48:03<21:51,  2.94s/it]
 83%|████████▎ | 2117/2562 [1:48:07<22:24,  3.02s/it]
 83%|████████▎ | 2118/2562 [1:48:10<22:10,  3.00s/it]
 83%|████████▎ | 2119/2562 [1:48:13<22:05,  2.99s/it]
 83%|████████▎ | 2120/2562 [1:48:16<21:58,  2.98s/it]
                                                     
{'loss': 0.7639, 'grad_norm': 4.84805154800415, 'learning_rate': 1.584927059781548e-06, 'epoch': 2.48}

 83%|████████▎ | 2120/2562 [1:48:16<21:58,  2.98s/it]
 83%|████████▎ | 2121/2562 [1:48:19<22:02,  3.00s/it]
 83%|████████▎ | 2122/2562 [1:48:21<21:45,  2.97s/it]
 83%|████████▎ | 2123/2562 [1:48:24<21:49,  2.98s/it]
 83%|████████▎ | 2124/2562 [1:48:27<21:21,  2.93s/it]
 83%|████████▎ | 2125/2562 [1:48:30<21:26,  2.94s/it]
 83%|████████▎ | 2126/2562 [1:48:33<21:23,  2.94s/it]
 83%|████████▎ | 2127/2562 [1:48:36<20:53,  2.88s/it]
 83%|████████▎ | 2128/2562 [1:48:39<20:44,  2.87s/it]
 83%|████████▎ | 2129/2562 [1:48:42<21:09,  2.93s/it]
 83%|████████▎ | 2130/2562 [1:48:45<21:02,  2.92s/it]
                                                     
{'loss': 0.7833, 'grad_norm': 4.458873271942139, 'learning_rate': 1.5158716939533524e-06, 'epoch': 2.49}

 83%|████████▎ | 2130/2562 [1:48:45<21:02,  2.92s/it]
 83%|████████▎ | 2131/2562 [1:48:48<20:57,  2.92s/it]
 83%|████████▎ | 2132/2562 [1:48:51<21:03,  2.94s/it]
 83%|████████▎ | 2133/2562 [1:48:54<20:50,  2.91s/it]
 83%|████████▎ | 2134/2562 [1:48:56<20:45,  2.91s/it]
 83%|████████▎ | 2135/2562 [1:48:59<20:45,  2.92s/it]
 83%|████████▎ | 2136/2562 [1:49:02<20:33,  2.90s/it]
 83%|████████▎ | 2137/2562 [1:49:05<20:28,  2.89s/it]
 83%|████████▎ | 2138/2562 [1:49:08<20:38,  2.92s/it]
 83%|████████▎ | 2139/2562 [1:49:11<20:23,  2.89s/it]
 84%|████████▎ | 2140/2562 [1:49:14<20:43,  2.95s/it]
                                                     
{'loss': 0.7962, 'grad_norm': 4.67988920211792, 'learning_rate': 1.448230872972568e-06, 'epoch': 2.51}

 84%|████████▎ | 2140/2562 [1:49:14<20:43,  2.95s/it]
 84%|████████▎ | 2141/2562 [1:49:17<21:05,  3.00s/it]
 84%|████████▎ | 2142/2562 [1:49:20<20:25,  2.92s/it]
 84%|████████▎ | 2143/2562 [1:49:23<20:21,  2.91s/it]
 84%|████████▎ | 2144/2562 [1:49:26<20:31,  2.95s/it]
 84%|████████▎ | 2145/2562 [1:49:29<20:07,  2.90s/it]
 84%|████████▍ | 2146/2562 [1:49:31<19:47,  2.86s/it]
 84%|████████▍ | 2147/2562 [1:49:34<19:57,  2.89s/it]
 84%|████████▍ | 2148/2562 [1:49:37<19:54,  2.89s/it]
 84%|████████▍ | 2149/2562 [1:49:40<19:50,  2.88s/it]
 84%|████████▍ | 2150/2562 [1:49:43<19:55,  2.90s/it]
                                                     
{'loss': 0.7777, 'grad_norm': 4.588193416595459, 'learning_rate': 1.382015874482735e-06, 'epoch': 2.52}

 84%|████████▍ | 2150/2562 [1:49:43<19:55,  2.90s/it]
 84%|████████▍ | 2151/2562 [1:49:46<19:41,  2.88s/it]
 84%|████████▍ | 2152/2562 [1:49:49<20:01,  2.93s/it]
 84%|████████▍ | 2153/2562 [1:49:52<19:52,  2.91s/it]
 84%|████████▍ | 2154/2562 [1:49:55<20:09,  2.96s/it]
 84%|████████▍ | 2155/2562 [1:49:58<20:02,  2.95s/it]
 84%|████████▍ | 2156/2562 [1:50:01<19:39,  2.90s/it]
 84%|████████▍ | 2157/2562 [1:50:03<19:32,  2.89s/it]
 84%|████████▍ | 2158/2562 [1:50:06<19:46,  2.94s/it]
 84%|████████▍ | 2159/2562 [1:50:09<19:28,  2.90s/it]
 84%|████████▍ | 2160/2562 [1:50:12<19:35,  2.92s/it]
                                                     
{'loss': 0.7748, 'grad_norm': 4.775571346282959, 'learning_rate': 1.3172377384023393e-06, 'epoch': 2.53}

 84%|████████▍ | 2160/2562 [1:50:12<19:35,  2.92s/it]
 84%|████████▍ | 2161/2562 [1:50:15<19:46,  2.96s/it]
 84%|████████▍ | 2162/2562 [1:50:18<19:27,  2.92s/it]
 84%|████████▍ | 2163/2562 [1:50:21<19:14,  2.89s/it]
 84%|████████▍ | 2164/2562 [1:50:24<19:17,  2.91s/it]
 85%|████████▍ | 2165/2562 [1:50:27<18:59,  2.87s/it]
 85%|████████▍ | 2166/2562 [1:50:29<18:53,  2.86s/it]
 85%|████████▍ | 2167/2562 [1:50:32<19:02,  2.89s/it]
 85%|████████▍ | 2168/2562 [1:50:35<18:45,  2.86s/it]
 85%|████████▍ | 2169/2562 [1:50:38<18:34,  2.84s/it]
 85%|████████▍ | 2170/2562 [1:50:41<18:49,  2.88s/it]
                                                     
{'loss': 0.7678, 'grad_norm': 4.795689105987549, 'learning_rate': 1.2539072650841523e-06, 'epoch': 2.54}

 85%|████████▍ | 2170/2562 [1:50:41<18:49,  2.88s/it]
 85%|████████▍ | 2171/2562 [1:50:44<19:09,  2.94s/it]
 85%|████████▍ | 2172/2562 [1:50:47<19:25,  2.99s/it]
 85%|████████▍ | 2173/2562 [1:50:50<19:24,  2.99s/it]
 85%|████████▍ | 2174/2562 [1:50:53<19:02,  2.94s/it]
 85%|████████▍ | 2175/2562 [1:50:56<18:43,  2.90s/it]
 85%|████████▍ | 2176/2562 [1:50:59<19:09,  2.98s/it]
 85%|████████▍ | 2177/2562 [1:51:02<18:36,  2.90s/it]
 85%|████████▌ | 2178/2562 [1:51:05<18:41,  2.92s/it]
 85%|████████▌ | 2179/2562 [1:51:08<18:41,  2.93s/it]
 85%|████████▌ | 2180/2562 [1:51:10<18:29,  2.91s/it]
                                                     
{'loss': 0.7825, 'grad_norm': 5.257193088531494, 'learning_rate': 1.1920350135144898e-06, 'epoch': 2.55}

 85%|████████▌ | 2180/2562 [1:51:10<18:29,  2.91s/it]
 85%|████████▌ | 2181/2562 [1:51:13<18:09,  2.86s/it]
 85%|████████▌ | 2182/2562 [1:51:16<18:18,  2.89s/it]
 85%|████████▌ | 2183/2562 [1:51:19<18:07,  2.87s/it]
 85%|████████▌ | 2184/2562 [1:51:22<17:56,  2.85s/it]
 85%|████████▌ | 2185/2562 [1:51:25<18:03,  2.87s/it]
 85%|████████▌ | 2186/2562 [1:51:27<17:50,  2.85s/it]
 85%|████████▌ | 2187/2562 [1:51:30<17:38,  2.82s/it]
 85%|████████▌ | 2188/2562 [1:51:33<17:57,  2.88s/it]
 85%|████████▌ | 2189/2562 [1:51:36<17:47,  2.86s/it]
 85%|████████▌ | 2190/2562 [1:51:39<17:45,  2.86s/it]
                                                     
{'loss': 0.7952, 'grad_norm': 4.7159342765808105, 'learning_rate': 1.1316312995527424e-06, 'epoch': 2.56}

 85%|████████▌ | 2190/2562 [1:51:39<17:45,  2.86s/it]
 86%|████████▌ | 2191/2562 [1:51:42<17:59,  2.91s/it]
 86%|████████▌ | 2192/2562 [1:51:45<17:39,  2.86s/it]
 86%|████████▌ | 2193/2562 [1:51:48<17:31,  2.85s/it]
 86%|████████▌ | 2194/2562 [1:51:51<17:55,  2.92s/it]
 86%|████████▌ | 2195/2562 [1:51:53<17:41,  2.89s/it]
 86%|████████▌ | 2196/2562 [1:51:56<17:26,  2.86s/it]
 86%|████████▌ | 2197/2562 [1:51:59<17:34,  2.89s/it]
 86%|████████▌ | 2198/2562 [1:52:02<17:29,  2.88s/it]
 86%|████████▌ | 2199/2562 [1:52:05<17:13,  2.85s/it]
 86%|████████▌ | 2200/2562 [1:52:08<17:29,  2.90s/it]
                                                     
{'loss': 0.7895, 'grad_norm': 5.012643337249756, 'learning_rate': 1.072706194211426e-06, 'epoch': 2.58}

 86%|████████▌ | 2200/2562 [1:52:08<17:29,  2.90s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.99it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.84it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.27it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.94it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.57it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.75it/s]

 10%|█         | 14/140 [00:01<00:12,  9.80it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.84it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.73it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.63it/s]

 14%|█▍        | 20/140 [00:01<00:12,  9.70it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.52it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.46it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.57it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.43it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.13it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.17it/s]

 20%|██        | 28/140 [00:02<00:11,  9.47it/s]

 21%|██        | 29/140 [00:02<00:11,  9.32it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.02it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.03it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.23it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.70it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.80it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.83it/s]

 28%|██▊       | 39/140 [00:03<00:10,  9.94it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.93it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 30%|███       | 42/140 [00:04<00:09,  9.93it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.90it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.89it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.65it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.32it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.25it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.17it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.23it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.42it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.43it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.45it/s]

 40%|████      | 56/140 [00:05<00:08,  9.72it/s]

 41%|████      | 57/140 [00:05<00:08,  9.51it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.41it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.53it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.65it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.85it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.65it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.57it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.53it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.51it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.72it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.81it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.83it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.86it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.87it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.85it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.86it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.61it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.23it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.37it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.31it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.23it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.12it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.15it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.37it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.22it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.19it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.09it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.01it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.07it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.26it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.43it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.27it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.02it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.12it/s]

 69%|██████▉   | 97/140 [00:10<00:04,  9.37it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.27it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.34it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.51it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.33it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.40it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.53it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.75it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.63it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.50it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.58it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.64it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.66it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.83it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.83it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.53it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.54it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.32it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.30it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.41it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.12it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.09it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.27it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.23it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.38it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.29it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.08it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.86it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.61it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.73it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.06it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.53it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.68it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.55it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.56it/s]

100%|██████████| 140/140 [00:14<00:00,  8.22it/s]
                                                     

                                                 
{'eval_loss': 0.9022672772407532, 'eval_runtime': 15.1046, 'eval_samples_per_second': 18.537, 'eval_steps_per_second': 9.269, 'epoch': 2.58}

 86%|████████▌ | 2200/2562 [1:52:23<17:29,  2.90s/it]

100%|██████████| 140/140 [00:14<00:00,  8.22it/s]

                                                 
 86%|████████▌ | 2201/2562 [1:52:26<44:47,  7.44s/it]
 86%|████████▌ | 2202/2562 [1:52:29<36:15,  6.04s/it]
 86%|████████▌ | 2203/2562 [1:52:32<30:42,  5.13s/it]
 86%|████████▌ | 2204/2562 [1:52:35<26:45,  4.48s/it]
 86%|████████▌ | 2205/2562 [1:52:38<23:50,  4.01s/it]
 86%|████████▌ | 2206/2562 [1:52:41<21:59,  3.71s/it]
 86%|████████▌ | 2207/2562 [1:52:44<20:36,  3.48s/it]
 86%|████████▌ | 2208/2562 [1:52:46<19:21,  3.28s/it]
 86%|████████▌ | 2209/2562 [1:52:49<18:48,  3.20s/it]
 86%|████████▋ | 2210/2562 [1:52:52<18:08,  3.09s/it]
                                                     
{'loss': 0.7973, 'grad_norm': 4.862824440002441, 'learning_rate': 1.0152695219770558e-06, 'epoch': 2.59}

 86%|████████▋ | 2210/2562 [1:52:52<18:08,  3.09s/it]
 86%|████████▋ | 2211/2562 [1:52:55<17:46,  3.04s/it]
 86%|████████▋ | 2212/2562 [1:52:58<17:58,  3.08s/it]
 86%|████████▋ | 2213/2562 [1:53:01<17:39,  3.04s/it]
 86%|████████▋ | 2214/2562 [1:53:04<17:39,  3.05s/it]
 86%|████████▋ | 2215/2562 [1:53:07<17:50,  3.09s/it]
 86%|████████▋ | 2216/2562 [1:53:10<17:24,  3.02s/it]
 87%|████████▋ | 2217/2562 [1:53:13<16:49,  2.93s/it]
 87%|████████▋ | 2218/2562 [1:53:16<17:08,  2.99s/it]
 87%|████████▋ | 2219/2562 [1:53:19<17:07,  3.00s/it]
 87%|████████▋ | 2220/2562 [1:53:22<17:04,  2.99s/it]
                                                     
{'loss': 0.7705, 'grad_norm': 4.596837520599365, 'learning_rate': 9.593308591721274e-07, 'epoch': 2.6}

 87%|████████▋ | 2220/2562 [1:53:22<17:04,  2.99s/it]
 87%|████████▋ | 2221/2562 [1:53:25<16:44,  2.95s/it]
 87%|████████▋ | 2222/2562 [1:53:28<16:25,  2.90s/it]
 87%|████████▋ | 2223/2562 [1:53:31<16:46,  2.97s/it]
 87%|████████▋ | 2224/2562 [1:53:34<16:33,  2.94s/it]
 87%|████████▋ | 2225/2562 [1:53:37<16:19,  2.91s/it]
 87%|████████▋ | 2226/2562 [1:53:40<16:31,  2.95s/it]
 87%|████████▋ | 2227/2562 [1:53:42<16:16,  2.91s/it]
 87%|████████▋ | 2228/2562 [1:53:45<16:00,  2.88s/it]
 87%|████████▋ | 2229/2562 [1:53:48<16:29,  2.97s/it]
 87%|████████▋ | 2230/2562 [1:53:51<16:08,  2.92s/it]
                                                     
{'loss': 0.7919, 'grad_norm': 4.939295291900635, 'learning_rate': 9.048995323584764e-07, 'epoch': 2.61}

 87%|████████▋ | 2230/2562 [1:53:51<16:08,  2.92s/it]
 87%|████████▋ | 2231/2562 [1:53:54<15:59,  2.90s/it]
 87%|████████▋ | 2232/2562 [1:53:57<16:12,  2.95s/it]
 87%|████████▋ | 2233/2562 [1:54:00<16:02,  2.93s/it]
 87%|████████▋ | 2234/2562 [1:54:03<15:56,  2.91s/it]
 87%|████████▋ | 2235/2562 [1:54:06<16:16,  2.99s/it]
 87%|████████▋ | 2236/2562 [1:54:09<15:58,  2.94s/it]
 87%|████████▋ | 2237/2562 [1:54:12<15:38,  2.89s/it]
 87%|████████▋ | 2238/2562 [1:54:15<15:37,  2.89s/it]
 87%|████████▋ | 2239/2562 [1:54:17<15:25,  2.87s/it]
 87%|████████▋ | 2240/2562 [1:54:20<15:14,  2.84s/it]
                                                     
{'loss': 0.7791, 'grad_norm': 4.718730926513672, 'learning_rate': 8.519846167822665e-07, 'epoch': 2.62}

 87%|████████▋ | 2240/2562 [1:54:20<15:14,  2.84s/it]
 87%|████████▋ | 2241/2562 [1:54:23<15:35,  2.91s/it]
 88%|████████▊ | 2242/2562 [1:54:26<15:20,  2.88s/it]
 88%|████████▊ | 2243/2562 [1:54:29<15:13,  2.86s/it]
 88%|████████▊ | 2244/2562 [1:54:32<15:23,  2.90s/it]
 88%|████████▊ | 2245/2562 [1:54:35<15:21,  2.91s/it]
 88%|████████▊ | 2246/2562 [1:54:38<15:21,  2.92s/it]
 88%|████████▊ | 2247/2562 [1:54:41<15:30,  2.96s/it]
 88%|████████▊ | 2248/2562 [1:54:44<15:16,  2.92s/it]
 88%|████████▊ | 2249/2562 [1:54:47<15:09,  2.91s/it]
 88%|████████▊ | 2250/2562 [1:54:50<15:20,  2.95s/it]
                                                     
{'loss': 0.7801, 'grad_norm': 4.674426078796387, 'learning_rate': 8.005949348608977e-07, 'epoch': 2.63}

 88%|████████▊ | 2250/2562 [1:54:50<15:20,  2.95s/it]
 88%|████████▊ | 2251/2562 [1:54:52<15:11,  2.93s/it]
 88%|████████▊ | 2252/2562 [1:54:55<14:55,  2.89s/it]
 88%|████████▊ | 2253/2562 [1:54:58<15:07,  2.94s/it]
 88%|████████▊ | 2254/2562 [1:55:01<14:55,  2.91s/it]
 88%|████████▊ | 2255/2562 [1:55:04<14:52,  2.91s/it]
 88%|████████▊ | 2256/2562 [1:55:07<14:59,  2.94s/it]
 88%|████████▊ | 2257/2562 [1:55:10<14:38,  2.88s/it]
 88%|████████▊ | 2258/2562 [1:55:13<14:31,  2.87s/it]
 88%|████████▊ | 2259/2562 [1:55:16<14:59,  2.97s/it]
 88%|████████▊ | 2260/2562 [1:55:19<14:44,  2.93s/it]
                                                     
{'loss': 0.8052, 'grad_norm': 4.957338333129883, 'learning_rate': 7.507390547120541e-07, 'epoch': 2.65}

 88%|████████▊ | 2260/2562 [1:55:19<14:44,  2.93s/it]
 88%|████████▊ | 2261/2562 [1:55:21<14:29,  2.89s/it]
 88%|████████▊ | 2262/2562 [1:55:24<14:37,  2.93s/it]
 88%|████████▊ | 2263/2562 [1:55:27<14:21,  2.88s/it]
 88%|████████▊ | 2264/2562 [1:55:30<14:08,  2.85s/it]
 88%|████████▊ | 2265/2562 [1:55:33<14:21,  2.90s/it]
 88%|████████▊ | 2266/2562 [1:55:36<14:07,  2.86s/it]
 88%|████████▊ | 2267/2562 [1:55:39<14:01,  2.85s/it]
 89%|████████▊ | 2268/2562 [1:55:42<14:06,  2.88s/it]
 89%|████████▊ | 2269/2562 [1:55:44<14:04,  2.88s/it]
 89%|████████▊ | 2270/2562 [1:55:47<14:10,  2.91s/it]
                                                     
{'loss': 0.7758, 'grad_norm': 4.532878398895264, 'learning_rate': 7.024252887251548e-07, 'epoch': 2.66}

 89%|████████▊ | 2270/2562 [1:55:47<14:10,  2.91s/it]
 89%|████████▊ | 2271/2562 [1:55:51<14:22,  2.96s/it]
 89%|████████▊ | 2272/2562 [1:55:53<14:13,  2.94s/it]
 89%|████████▊ | 2273/2562 [1:55:56<14:02,  2.91s/it]
 89%|████████▉ | 2274/2562 [1:55:59<14:24,  3.00s/it]
 89%|████████▉ | 2275/2562 [1:56:02<14:02,  2.94s/it]
 89%|████████▉ | 2276/2562 [1:56:05<13:52,  2.91s/it]
 89%|████████▉ | 2277/2562 [1:56:08<13:51,  2.92s/it]
 89%|████████▉ | 2278/2562 [1:56:11<13:48,  2.92s/it]
 89%|████████▉ | 2279/2562 [1:56:14<13:39,  2.89s/it]
 89%|████████▉ | 2280/2562 [1:56:17<13:49,  2.94s/it]
                                                     
{'loss': 0.786, 'grad_norm': 4.901816368103027, 'learning_rate': 6.556616921754489e-07, 'epoch': 2.67}

 89%|████████▉ | 2280/2562 [1:56:17<13:49,  2.94s/it]
 89%|████████▉ | 2281/2562 [1:56:20<13:38,  2.91s/it]
 89%|████████▉ | 2282/2562 [1:56:23<13:26,  2.88s/it]
 89%|████████▉ | 2283/2562 [1:56:26<13:35,  2.92s/it]
 89%|████████▉ | 2284/2562 [1:56:29<13:39,  2.95s/it]
 89%|████████▉ | 2285/2562 [1:56:32<13:45,  2.98s/it]
 89%|████████▉ | 2286/2562 [1:56:35<13:40,  2.97s/it]
 89%|████████▉ | 2287/2562 [1:56:37<13:32,  2.95s/it]
 89%|████████▉ | 2288/2562 [1:56:40<13:27,  2.95s/it]
 89%|████████▉ | 2289/2562 [1:56:43<13:07,  2.88s/it]
 89%|████████▉ | 2290/2562 [1:56:46<12:59,  2.87s/it]
                                                     
{'loss': 0.7873, 'grad_norm': 4.98193359375, 'learning_rate': 6.10456061880963e-07, 'epoch': 2.68}

 89%|████████▉ | 2290/2562 [1:56:46<12:59,  2.87s/it]
 89%|████████▉ | 2291/2562 [1:56:49<13:01,  2.88s/it]
 89%|████████▉ | 2292/2562 [1:56:52<12:51,  2.86s/it]
 90%|████████▉ | 2293/2562 [1:56:55<12:51,  2.87s/it]
 90%|████████▉ | 2294/2562 [1:56:58<13:09,  2.94s/it]
 90%|████████▉ | 2295/2562 [1:57:01<13:01,  2.93s/it]
 90%|████████▉ | 2296/2562 [1:57:04<13:12,  2.98s/it]
 90%|████████▉ | 2297/2562 [1:57:06<12:53,  2.92s/it]
 90%|████████▉ | 2298/2562 [1:57:09<12:49,  2.91s/it]
 90%|████████▉ | 2299/2562 [1:57:12<12:52,  2.94s/it]
 90%|████████▉ | 2300/2562 [1:57:15<12:58,  2.97s/it]
                                                     
{'loss': 0.7898, 'grad_norm': 4.918573379516602, 'learning_rate': 5.668159349025649e-07, 'epoch': 2.69}

 90%|████████▉ | 2300/2562 [1:57:15<12:58,  2.97s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.79it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.73it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.20it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.86it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.53it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.70it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.75it/s]

 11%|█         | 15/140 [00:01<00:12,  9.79it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.77it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.82it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.59it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.48it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.62it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.39it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.36it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.50it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.33it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.03it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.11it/s]

 20%|██        | 28/140 [00:02<00:11,  9.41it/s]

 21%|██        | 29/140 [00:02<00:11,  9.27it/s]

 21%|██▏       | 30/140 [00:03<00:12,  8.97it/s]

 22%|██▏       | 31/140 [00:03<00:12,  8.99it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.21it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.67it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.79it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.82it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.95it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.92it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 30%|███       | 42/140 [00:04<00:09,  9.93it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.93it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.89it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.65it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.32it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.28it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.18it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.22it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.41it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.40it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.41it/s]

 40%|████      | 56/140 [00:05<00:08,  9.70it/s]

 41%|████      | 57/140 [00:05<00:08,  9.49it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.39it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.51it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.61it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.79it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.59it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.54it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.51it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.48it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.70it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.73it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.78it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.78it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.83it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.85it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.82it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.81it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.56it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.18it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.33it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.28it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.21it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.11it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.12it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.32it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.20it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.16it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.05it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.08it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  8.98it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.07it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.25it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.42it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.26it/s]

 68%|██████▊   | 95/140 [00:09<00:05,  8.98it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.09it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.28it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.32it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.45it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.29it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.34it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.51it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.75it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.63it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.52it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.59it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.66it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.68it/s]

 80%|████████  | 112/140 [00:11<00:02,  9.76it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.85it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.54it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.52it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.30it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.28it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.39it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.10it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.07it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.24it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.22it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.38it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.27it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.06it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.83it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.58it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.69it/s]

 95%|█████████▌| 133/140 [00:14<00:00,  9.01it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.27it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.63it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.67it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.47it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.49it/s]

100%|██████████| 140/140 [00:14<00:00,  8.07it/s]
                                                     

                                                 
{'eval_loss': 0.8991779685020447, 'eval_runtime': 15.1523, 'eval_samples_per_second': 18.479, 'eval_steps_per_second': 9.24, 'epoch': 2.69}

 90%|████████▉ | 2300/2562 [1:57:31<12:58,  2.97s/it]

100%|██████████| 140/140 [00:14<00:00,  8.07it/s]

                                                 
 90%|████████▉ | 2301/2562 [1:57:33<32:36,  7.49s/it]
 90%|████████▉ | 2302/2562 [1:57:36<26:36,  6.14s/it]
 90%|████████▉ | 2303/2562 [1:57:39<22:21,  5.18s/it]
 90%|████████▉ | 2304/2562 [1:57:42<19:14,  4.47s/it]
 90%|████████▉ | 2305/2562 [1:57:45<17:06,  3.99s/it]
 90%|█████████ | 2306/2562 [1:57:48<15:50,  3.71s/it]
 90%|█████████ | 2307/2562 [1:57:51<14:55,  3.51s/it]
 90%|█████████ | 2308/2562 [1:57:54<14:24,  3.40s/it]
 90%|█████████ | 2309/2562 [1:57:57<13:53,  3.30s/it]
 90%|█████████ | 2310/2562 [1:58:00<13:20,  3.18s/it]
                                                     
{'loss': 0.8007, 'grad_norm': 4.899518013000488, 'learning_rate': 5.247485872873026e-07, 'epoch': 2.7}

 90%|█████████ | 2310/2562 [1:58:00<13:20,  3.18s/it]
 90%|█████████ | 2311/2562 [1:58:03<13:03,  3.12s/it]
 90%|█████████ | 2312/2562 [1:58:06<12:49,  3.08s/it]
 90%|█████████ | 2313/2562 [1:58:09<12:36,  3.04s/it]
 90%|█████████ | 2314/2562 [1:58:12<12:40,  3.07s/it]
 90%|█████████ | 2315/2562 [1:58:15<12:17,  2.98s/it]
 90%|█████████ | 2316/2562 [1:58:18<12:30,  3.05s/it]
 90%|█████████ | 2317/2562 [1:58:21<12:24,  3.04s/it]
 90%|█████████ | 2318/2562 [1:58:24<12:11,  3.00s/it]
 91%|█████████ | 2319/2562 [1:58:27<11:54,  2.94s/it]
 91%|█████████ | 2320/2562 [1:58:30<12:11,  3.02s/it]
                                                     
{'loss': 0.7813, 'grad_norm': 4.619868755340576, 'learning_rate': 4.842610328552999e-07, 'epoch': 2.72}

 91%|█████████ | 2320/2562 [1:58:30<12:11,  3.02s/it]
 91%|█████████ | 2321/2562 [1:58:33<11:59,  2.98s/it]
 91%|█████████ | 2322/2562 [1:58:36<11:55,  2.98s/it]
 91%|█████████ | 2323/2562 [1:58:39<12:05,  3.04s/it]
 91%|█████████ | 2324/2562 [1:58:42<11:41,  2.95s/it]
 91%|█████████ | 2325/2562 [1:58:45<11:34,  2.93s/it]
 91%|█████████ | 2326/2562 [1:58:48<11:30,  2.92s/it]
 91%|█████████ | 2327/2562 [1:58:51<11:23,  2.91s/it]
 91%|█████████ | 2328/2562 [1:58:53<11:10,  2.87s/it]
 91%|█████████ | 2329/2562 [1:58:56<11:14,  2.89s/it]
 91%|█████████ | 2330/2562 [1:58:59<11:06,  2.87s/it]
                                                     
{'loss': 0.7937, 'grad_norm': 4.746974945068359, 'learning_rate': 4.453600220303378e-07, 'epoch': 2.73}

 91%|█████████ | 2330/2562 [1:58:59<11:06,  2.87s/it]
 91%|█████████ | 2331/2562 [1:59:02<11:15,  2.93s/it]
 91%|█████████ | 2332/2562 [1:59:05<11:09,  2.91s/it]
 91%|█████████ | 2333/2562 [1:59:08<10:59,  2.88s/it]
 91%|█████████ | 2334/2562 [1:59:11<11:05,  2.92s/it]
 91%|█████████ | 2335/2562 [1:59:14<11:00,  2.91s/it]
 91%|█████████ | 2336/2562 [1:59:17<10:50,  2.88s/it]
 91%|█████████ | 2337/2562 [1:59:20<10:51,  2.90s/it]
 91%|█████████▏| 2338/2562 [1:59:23<10:51,  2.91s/it]
 91%|█████████▏| 2339/2562 [1:59:25<10:39,  2.87s/it]
 91%|█████████▏| 2340/2562 [1:59:28<10:37,  2.87s/it]
                                                     
{'loss': 0.8093, 'grad_norm': 4.999614238739014, 'learning_rate': 4.0805204071437953e-07, 'epoch': 2.74}

 91%|█████████▏| 2340/2562 [1:59:28<10:37,  2.87s/it]
 91%|█████████▏| 2341/2562 [1:59:31<10:31,  2.86s/it]
 91%|█████████▏| 2342/2562 [1:59:34<10:18,  2.81s/it]
 91%|█████████▏| 2343/2562 [1:59:37<10:24,  2.85s/it]
 91%|█████████▏| 2344/2562 [1:59:40<10:29,  2.89s/it]
 92%|█████████▏| 2345/2562 [1:59:42<10:23,  2.87s/it]
 92%|█████████▏| 2346/2562 [1:59:46<10:32,  2.93s/it]
 92%|█████████▏| 2347/2562 [1:59:48<10:27,  2.92s/it]
 92%|█████████▏| 2348/2562 [1:59:51<10:23,  2.91s/it]
 92%|█████████▏| 2349/2562 [1:59:54<10:23,  2.93s/it]
 92%|█████████▏| 2350/2562 [1:59:57<10:23,  2.94s/it]
                                                     
{'loss': 0.7645, 'grad_norm': 4.524498462677002, 'learning_rate': 3.72343309206179e-07, 'epoch': 2.75}

 92%|█████████▏| 2350/2562 [1:59:57<10:23,  2.94s/it]
 92%|█████████▏| 2351/2562 [2:00:00<10:14,  2.91s/it]
 92%|█████████▏| 2352/2562 [2:00:03<10:14,  2.93s/it]
 92%|█████████▏| 2353/2562 [2:00:06<10:02,  2.88s/it]
 92%|█████████▏| 2354/2562 [2:00:09<10:00,  2.89s/it]
 92%|█████████▏| 2355/2562 [2:00:12<10:06,  2.93s/it]
 92%|█████████▏| 2356/2562 [2:00:15<10:10,  2.96s/it]
 92%|█████████▏| 2357/2562 [2:00:18<10:29,  3.07s/it]
 92%|█████████▏| 2358/2562 [2:00:21<10:18,  3.03s/it]
 92%|█████████▏| 2359/2562 [2:00:24<10:06,  2.99s/it]
 92%|█████████▏| 2360/2562 [2:00:27<10:10,  3.02s/it]
                                                     
{'loss': 0.7706, 'grad_norm': 5.050011157989502, 'learning_rate': 3.382397811641858e-07, 'epoch': 2.76}

 92%|█████████▏| 2360/2562 [2:00:27<10:10,  3.02s/it]
 92%|█████████▏| 2361/2562 [2:00:30<10:03,  3.00s/it]
 92%|█████████▏| 2362/2562 [2:00:33<09:53,  2.97s/it]
 92%|█████████▏| 2363/2562 [2:00:36<09:52,  2.98s/it]
 92%|█████████▏| 2364/2562 [2:00:39<09:36,  2.91s/it]
 92%|█████████▏| 2365/2562 [2:00:42<09:30,  2.89s/it]
 92%|█████████▏| 2366/2562 [2:00:45<09:41,  2.97s/it]
 92%|█████████▏| 2367/2562 [2:00:47<09:25,  2.90s/it]
 92%|█████████▏| 2368/2562 [2:00:50<09:20,  2.89s/it]
 92%|█████████▏| 2369/2562 [2:00:53<09:19,  2.90s/it]
 93%|█████████▎| 2370/2562 [2:00:56<09:05,  2.84s/it]
                                                     
{'loss': 0.778, 'grad_norm': 4.953895092010498, 'learning_rate': 3.057471426138958e-07, 'epoch': 2.77}

 93%|█████████▎| 2370/2562 [2:00:56<09:05,  2.84s/it]
 93%|█████████▎| 2371/2562 [2:00:59<08:59,  2.82s/it]
 93%|█████████▎| 2372/2562 [2:01:02<08:59,  2.84s/it]
 93%|█████████▎| 2373/2562 [2:01:05<09:02,  2.87s/it]
 93%|█████████▎| 2374/2562 [2:01:07<08:54,  2.84s/it]
 93%|█████████▎| 2375/2562 [2:01:10<09:04,  2.91s/it]
 93%|█████████▎| 2376/2562 [2:01:13<08:51,  2.86s/it]
 93%|█████████▎| 2377/2562 [2:01:16<08:46,  2.85s/it]
 93%|█████████▎| 2378/2562 [2:01:19<08:59,  2.93s/it]
 93%|█████████▎| 2379/2562 [2:01:22<08:57,  2.94s/it]
 93%|█████████▎| 2380/2562 [2:01:25<08:48,  2.90s/it]
                                                     
{'loss': 0.7855, 'grad_norm': 5.001157283782959, 'learning_rate': 2.7487081099983435e-07, 'epoch': 2.79}

 93%|█████████▎| 2380/2562 [2:01:25<08:48,  2.90s/it]
 93%|█████████▎| 2381/2562 [2:01:28<08:49,  2.93s/it]
 93%|█████████▎| 2382/2562 [2:01:31<08:44,  2.91s/it]
 93%|█████████▎| 2383/2562 [2:01:34<08:36,  2.88s/it]
 93%|█████████▎| 2384/2562 [2:01:37<08:39,  2.92s/it]
 93%|█████████▎| 2385/2562 [2:01:40<08:41,  2.95s/it]
 93%|█████████▎| 2386/2562 [2:01:42<08:36,  2.94s/it]
 93%|█████████▎| 2387/2562 [2:01:45<08:33,  2.93s/it]
 93%|█████████▎| 2388/2562 [2:01:48<08:27,  2.92s/it]
 93%|█████████▎| 2389/2562 [2:01:51<08:34,  2.98s/it]
 93%|█████████▎| 2390/2562 [2:01:54<08:31,  2.97s/it]
                                                     
{'loss': 0.7898, 'grad_norm': 4.771422863006592, 'learning_rate': 2.4561593428231165e-07, 'epoch': 2.8}

 93%|█████████▎| 2390/2562 [2:01:54<08:31,  2.97s/it]
 93%|█████████▎| 2391/2562 [2:01:57<08:20,  2.93s/it]
 93%|█████████▎| 2392/2562 [2:02:00<08:26,  2.98s/it]
 93%|█████████▎| 2393/2562 [2:02:03<08:14,  2.93s/it]
 93%|█████████▎| 2394/2562 [2:02:06<08:14,  2.94s/it]
 93%|█████████▎| 2395/2562 [2:02:09<08:08,  2.92s/it]
 94%|█████████▎| 2396/2562 [2:02:12<08:07,  2.94s/it]
 94%|█████████▎| 2397/2562 [2:02:15<07:59,  2.91s/it]
 94%|█████████▎| 2398/2562 [2:02:18<07:56,  2.91s/it]
 94%|█████████▎| 2399/2562 [2:02:21<07:59,  2.94s/it]
 94%|█████████▎| 2400/2562 [2:02:23<07:51,  2.91s/it]
                                                     
{'loss': 0.7762, 'grad_norm': 5.319340705871582, 'learning_rate': 2.1798739007911517e-07, 'epoch': 2.81}

 94%|█████████▎| 2400/2562 [2:02:23<07:51,  2.91s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.56it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.71it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.21it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.89it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.57it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.74it/s]

 10%|█         | 14/140 [00:01<00:12,  9.78it/s]

 11%|█         | 15/140 [00:01<00:12,  9.80it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.78it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.82it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.63it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.56it/s]

 14%|█▍        | 20/140 [00:02<00:12,  9.67it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.43it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.38it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.51it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.34it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.05it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.14it/s]

 20%|██        | 28/140 [00:02<00:11,  9.50it/s]

 21%|██        | 29/140 [00:02<00:11,  9.36it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.06it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.06it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.27it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.73it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.83it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.85it/s]

 28%|██▊       | 39/140 [00:04<00:10,  9.94it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.93it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.94it/s]

 31%|███       | 43/140 [00:04<00:09,  9.98it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.96it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.91it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.64it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.30it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.27it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.18it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.20it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.41it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.45it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.46it/s]

 40%|████      | 56/140 [00:05<00:08,  9.71it/s]

 41%|████      | 57/140 [00:05<00:08,  9.51it/s]

 41%|████▏     | 58/140 [00:05<00:08,  9.41it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.53it/s]

 43%|████▎     | 60/140 [00:06<00:08,  9.61it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.82it/s]

 45%|████▌     | 63/140 [00:06<00:07,  9.64it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.57it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.55it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.50it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.72it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.78it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.79it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.80it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.87it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.85it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.83it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.63it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.24it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.37it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.32it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.24it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.14it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.16it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.34it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.22it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.21it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.10it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.13it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.02it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.08it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.24it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.41it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.27it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.03it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.11it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.28it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.34it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.48it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.32it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.36it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.52it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.74it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.63it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.53it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.62it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.65it/s]

 79%|███████▉  | 111/140 [00:11<00:02,  9.68it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.85it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.86it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.57it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.46it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.60it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.35it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.31it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.45it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.14it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.11it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.27it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.23it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.39it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.31it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.08it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.84it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.60it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.70it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.00it/s]

 96%|█████████▌| 134/140 [00:14<00:00,  9.28it/s]

 97%|█████████▋| 136/140 [00:14<00:00,  9.64it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.66it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.47it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.50it/s]

100%|██████████| 140/140 [00:14<00:00,  8.04it/s]
                                                     

                                                 
{'eval_loss': 0.8972675204277039, 'eval_runtime': 15.1112, 'eval_samples_per_second': 18.529, 'eval_steps_per_second': 9.265, 'epoch': 2.81}

 94%|█████████▎| 2400/2562 [2:02:39<07:51,  2.91s/it]

100%|██████████| 140/140 [00:14<00:00,  8.04it/s]

                                                 
 94%|█████████▎| 2401/2562 [2:02:42<20:07,  7.50s/it]
 94%|█████████▍| 2402/2562 [2:02:45<16:21,  6.13s/it]
 94%|█████████▍| 2403/2562 [2:02:48<13:44,  5.18s/it]
 94%|█████████▍| 2404/2562 [2:02:51<11:59,  4.55s/it]
 94%|█████████▍| 2405/2562 [2:02:53<10:30,  4.02s/it]
 94%|█████████▍| 2406/2562 [2:02:56<09:31,  3.66s/it]
 94%|█████████▍| 2407/2562 [2:02:59<08:59,  3.48s/it]
 94%|█████████▍| 2408/2562 [2:03:02<08:28,  3.30s/it]
 94%|█████████▍| 2409/2562 [2:03:05<08:03,  3.16s/it]
 94%|█████████▍| 2410/2562 [2:03:08<07:57,  3.14s/it]
                                                     
{'loss': 0.8041, 'grad_norm': 4.7191362380981445, 'learning_rate': 1.919897848522656e-07, 'epoch': 2.82}

 94%|█████████▍| 2410/2562 [2:03:08<07:57,  3.14s/it]
 94%|█████████▍| 2411/2562 [2:03:11<07:37,  3.03s/it]
 94%|█████████▍| 2412/2562 [2:03:14<07:35,  3.04s/it]
 94%|█████████▍| 2413/2562 [2:03:17<07:29,  3.02s/it]
 94%|█████████▍| 2414/2562 [2:03:20<07:13,  2.93s/it]
 94%|█████████▍| 2415/2562 [2:03:23<07:11,  2.93s/it]
 94%|█████████▍| 2416/2562 [2:03:26<07:13,  2.97s/it]
 94%|█████████▍| 2417/2562 [2:03:29<07:08,  2.95s/it]
 94%|█████████▍| 2418/2562 [2:03:31<06:57,  2.90s/it]
 94%|█████████▍| 2419/2562 [2:03:34<07:04,  2.97s/it]
 94%|█████████▍| 2420/2562 [2:03:37<06:58,  2.95s/it]
                                                     
{'loss': 0.807, 'grad_norm': 5.232530117034912, 'learning_rate': 1.6762745313999795e-07, 'epoch': 2.83}

 94%|█████████▍| 2420/2562 [2:03:37<06:58,  2.95s/it]
 94%|█████████▍| 2421/2562 [2:03:40<06:53,  2.93s/it]
 95%|█████████▍| 2422/2562 [2:03:43<06:47,  2.91s/it]
 95%|█████████▍| 2423/2562 [2:03:46<06:48,  2.94s/it]
 95%|█████████▍| 2424/2562 [2:03:49<06:41,  2.91s/it]
 95%|█████████▍| 2425/2562 [2:03:52<06:34,  2.88s/it]
 95%|█████████▍| 2426/2562 [2:03:55<06:31,  2.88s/it]
 95%|█████████▍| 2427/2562 [2:03:58<06:34,  2.92s/it]
 95%|█████████▍| 2428/2562 [2:04:01<06:30,  2.91s/it]
 95%|█████████▍| 2429/2562 [2:04:03<06:27,  2.91s/it]
 95%|█████████▍| 2430/2562 [2:04:06<06:27,  2.94s/it]
                                                     
{'loss': 0.8024, 'grad_norm': 4.943942070007324, 'learning_rate': 1.449044568340663e-07, 'epoch': 2.84}

 95%|█████████▍| 2430/2562 [2:04:06<06:27,  2.94s/it]
 95%|█████████▍| 2431/2562 [2:04:10<06:28,  2.97s/it]
 95%|█████████▍| 2432/2562 [2:04:12<06:20,  2.93s/it]
 95%|█████████▍| 2433/2562 [2:04:15<06:19,  2.95s/it]
 95%|█████████▌| 2434/2562 [2:04:18<06:12,  2.91s/it]
 95%|█████████▌| 2435/2562 [2:04:21<06:04,  2.87s/it]
 95%|█████████▌| 2436/2562 [2:04:24<06:12,  2.95s/it]
 95%|█████████▌| 2437/2562 [2:04:27<06:07,  2.94s/it]
 95%|█████████▌| 2438/2562 [2:04:30<06:03,  2.93s/it]
 95%|█████████▌| 2439/2562 [2:04:33<06:00,  2.93s/it]
 95%|█████████▌| 2440/2562 [2:04:36<05:53,  2.89s/it]
                                                     
{'loss': 0.8042, 'grad_norm': 4.9606523513793945, 'learning_rate': 1.2382458450250657e-07, 'epoch': 2.86}

 95%|█████████▌| 2440/2562 [2:04:36<05:53,  2.89s/it]
 95%|█████████▌| 2441/2562 [2:04:38<05:46,  2.86s/it]
 95%|█████████▌| 2442/2562 [2:04:41<05:48,  2.90s/it]
 95%|█████████▌| 2443/2562 [2:04:44<05:42,  2.88s/it]
 95%|█████████▌| 2444/2562 [2:04:47<05:39,  2.88s/it]
 95%|█████████▌| 2445/2562 [2:04:50<05:47,  2.97s/it]
 95%|█████████▌| 2446/2562 [2:04:53<05:39,  2.93s/it]
 96%|█████████▌| 2447/2562 [2:04:56<05:29,  2.86s/it]
 96%|█████████▌| 2448/2562 [2:04:59<05:29,  2.89s/it]
 96%|█████████▌| 2449/2562 [2:05:02<05:24,  2.87s/it]
 96%|█████████▌| 2450/2562 [2:05:05<05:22,  2.88s/it]
                                                     
{'loss': 0.7841, 'grad_norm': 5.00095796585083, 'learning_rate': 1.0439135075798634e-07, 'epoch': 2.87}

 96%|█████████▌| 2450/2562 [2:05:05<05:22,  2.88s/it]
 96%|█████████▌| 2451/2562 [2:05:08<05:26,  2.94s/it]
 96%|█████████▌| 2452/2562 [2:05:11<05:26,  2.97s/it]
 96%|█████████▌| 2453/2562 [2:05:13<05:18,  2.92s/it]
 96%|█████████▌| 2454/2562 [2:05:16<05:17,  2.94s/it]
 96%|█████████▌| 2455/2562 [2:05:19<05:10,  2.91s/it]
 96%|█████████▌| 2456/2562 [2:05:22<05:02,  2.85s/it]
 96%|█████████▌| 2457/2562 [2:05:25<05:02,  2.88s/it]
 96%|█████████▌| 2458/2562 [2:05:28<04:56,  2.85s/it]
 96%|█████████▌| 2459/2562 [2:05:30<04:49,  2.81s/it]
 96%|█████████▌| 2460/2562 [2:05:33<04:51,  2.86s/it]
                                                     
{'loss': 0.7655, 'grad_norm': 4.9865264892578125, 'learning_rate': 8.6607995671808e-08, 'epoch': 2.88}

 96%|█████████▌| 2460/2562 [2:05:33<04:51,  2.86s/it]
 96%|█████████▌| 2461/2562 [2:05:36<04:47,  2.85s/it]
 96%|█████████▌| 2462/2562 [2:05:39<04:50,  2.90s/it]
 96%|█████████▌| 2463/2562 [2:05:42<04:48,  2.91s/it]
 96%|█████████▌| 2464/2562 [2:05:45<04:47,  2.93s/it]
 96%|█████████▌| 2465/2562 [2:05:48<04:45,  2.94s/it]
 96%|█████████▋| 2466/2562 [2:05:51<04:45,  2.98s/it]
 96%|█████████▋| 2467/2562 [2:05:54<04:36,  2.91s/it]
 96%|█████████▋| 2468/2562 [2:05:57<04:31,  2.89s/it]
 96%|█████████▋| 2469/2562 [2:06:00<04:33,  2.94s/it]
 96%|█████████▋| 2470/2562 [2:06:03<04:28,  2.91s/it]
                                                     
{'loss': 0.7665, 'grad_norm': 4.681450843811035, 'learning_rate': 7.047748423370193e-08, 'epoch': 2.89}

 96%|█████████▋| 2470/2562 [2:06:03<04:28,  2.91s/it]
 96%|█████████▋| 2471/2562 [2:06:06<04:27,  2.94s/it]
 96%|█████████▋| 2472/2562 [2:06:08<04:19,  2.88s/it]
 97%|█████████▋| 2473/2562 [2:06:11<04:15,  2.87s/it]
 97%|█████████▋| 2474/2562 [2:06:14<04:16,  2.91s/it]
 97%|█████████▋| 2475/2562 [2:06:17<04:14,  2.92s/it]
 97%|█████████▋| 2476/2562 [2:06:20<04:10,  2.91s/it]
 97%|█████████▋| 2477/2562 [2:06:23<04:07,  2.91s/it]
 97%|█████████▋| 2478/2562 [2:06:26<04:05,  2.92s/it]
 97%|█████████▋| 2479/2562 [2:06:29<04:02,  2.92s/it]
 97%|█████████▋| 2480/2562 [2:06:32<03:59,  2.92s/it]
                                                     
{'loss': 0.7601, 'grad_norm': 4.992751121520996, 'learning_rate': 5.6002505857480906e-08, 'epoch': 2.9}

 97%|█████████▋| 2480/2562 [2:06:32<03:59,  2.92s/it]
 97%|█████████▋| 2481/2562 [2:06:35<03:54,  2.89s/it]
 97%|█████████▋| 2482/2562 [2:06:38<03:54,  2.93s/it]
 97%|█████████▋| 2483/2562 [2:06:41<03:56,  2.99s/it]
 97%|█████████▋| 2484/2562 [2:06:44<03:48,  2.92s/it]
 97%|█████████▋| 2485/2562 [2:06:46<03:41,  2.87s/it]
 97%|█████████▋| 2486/2562 [2:06:49<03:42,  2.93s/it]
 97%|█████████▋| 2487/2562 [2:06:52<03:36,  2.89s/it]
 97%|█████████▋| 2488/2562 [2:06:55<03:35,  2.91s/it]
 97%|█████████▋| 2489/2562 [2:06:58<03:36,  2.96s/it]
 97%|█████████▋| 2490/2562 [2:07:01<03:31,  2.94s/it]
                                                     
{'loss': 0.7695, 'grad_norm': 4.891352653503418, 'learning_rate': 4.318547393263317e-08, 'epoch': 2.91}

 97%|█████████▋| 2490/2562 [2:07:01<03:31,  2.94s/it]
 97%|█████████▋| 2491/2562 [2:07:04<03:28,  2.94s/it]
 97%|█████████▋| 2492/2562 [2:07:07<03:26,  2.95s/it]
 97%|█████████▋| 2493/2562 [2:07:10<03:20,  2.91s/it]
 97%|█████████▋| 2494/2562 [2:07:13<03:15,  2.88s/it]
 97%|█████████▋| 2495/2562 [2:07:16<03:14,  2.90s/it]
 97%|█████████▋| 2496/2562 [2:07:19<03:11,  2.90s/it]
 97%|█████████▋| 2497/2562 [2:07:21<03:08,  2.90s/it]
 98%|█████████▊| 2498/2562 [2:07:24<03:07,  2.92s/it]
 98%|█████████▊| 2499/2562 [2:07:27<03:04,  2.93s/it]
 98%|█████████▊| 2500/2562 [2:07:30<03:01,  2.93s/it]
                                                     
{'loss': 0.7668, 'grad_norm': 4.8527631759643555, 'learning_rate': 3.2028525421946563e-08, 'epoch': 2.93}

 98%|█████████▊| 2500/2562 [2:07:30<03:01,  2.93s/it]

  0%|          | 0/140 [00:00<?, ?it/s]

  1%|▏         | 2/140 [00:00<00:07, 18.64it/s]

  3%|▎         | 4/140 [00:00<00:11, 11.72it/s]

  4%|▍         | 6/140 [00:00<00:13, 10.19it/s]

  6%|▌         | 8/140 [00:00<00:13,  9.86it/s]

  7%|▋         | 10/140 [00:00<00:13,  9.53it/s]

  9%|▊         | 12/140 [00:01<00:13,  9.71it/s]

  9%|▉         | 13/140 [00:01<00:13,  9.73it/s]

 10%|█         | 14/140 [00:01<00:12,  9.77it/s]

 11%|█         | 15/140 [00:01<00:12,  9.79it/s]

 11%|█▏        | 16/140 [00:01<00:12,  9.77it/s]

 12%|█▏        | 17/140 [00:01<00:12,  9.80it/s]

 13%|█▎        | 18/140 [00:01<00:12,  9.56it/s]

 14%|█▎        | 19/140 [00:01<00:12,  9.51it/s]

 15%|█▌        | 21/140 [00:02<00:12,  9.47it/s]

 16%|█▌        | 22/140 [00:02<00:12,  9.43it/s]

 16%|█▋        | 23/140 [00:02<00:12,  9.54it/s]

 17%|█▋        | 24/140 [00:02<00:12,  9.41it/s]

 18%|█▊        | 25/140 [00:02<00:12,  9.12it/s]

 19%|█▊        | 26/140 [00:02<00:12,  9.18it/s]

 20%|██        | 28/140 [00:02<00:11,  9.50it/s]

 21%|██        | 29/140 [00:02<00:11,  9.34it/s]

 21%|██▏       | 30/140 [00:03<00:12,  9.03it/s]

 22%|██▏       | 31/140 [00:03<00:12,  9.04it/s]

 23%|██▎       | 32/140 [00:03<00:11,  9.24it/s]

 24%|██▍       | 34/140 [00:03<00:10,  9.68it/s]

 25%|██▌       | 35/140 [00:03<00:10,  9.75it/s]

 26%|██▌       | 36/140 [00:03<00:10,  9.77it/s]

 26%|██▋       | 37/140 [00:03<00:10,  9.81it/s]

 27%|██▋       | 38/140 [00:03<00:10,  9.86it/s]

 29%|██▊       | 40/140 [00:04<00:10,  9.94it/s]

 29%|██▉       | 41/140 [00:04<00:09,  9.92it/s]

 31%|███       | 43/140 [00:04<00:09,  9.97it/s]

 31%|███▏      | 44/140 [00:04<00:09,  9.94it/s]

 32%|███▏      | 45/140 [00:04<00:09,  9.92it/s]

 33%|███▎      | 46/140 [00:04<00:09,  9.66it/s]

 34%|███▎      | 47/140 [00:04<00:09,  9.32it/s]

 34%|███▍      | 48/140 [00:04<00:09,  9.27it/s]

 35%|███▌      | 49/140 [00:05<00:09,  9.18it/s]

 36%|███▌      | 50/140 [00:05<00:09,  9.24it/s]

 36%|███▋      | 51/140 [00:05<00:09,  9.43it/s]

 38%|███▊      | 53/140 [00:05<00:09,  9.47it/s]

 39%|███▊      | 54/140 [00:05<00:09,  9.46it/s]

 40%|████      | 56/140 [00:05<00:08,  9.73it/s]

 41%|████      | 57/140 [00:05<00:08,  9.50it/s]

 41%|████▏     | 58/140 [00:06<00:08,  9.41it/s]

 42%|████▏     | 59/140 [00:06<00:08,  9.52it/s]

 44%|████▎     | 61/140 [00:06<00:08,  9.77it/s]

 44%|████▍     | 62/140 [00:06<00:07,  9.81it/s]

 45%|████▌     | 63/140 [00:06<00:08,  9.61it/s]

 46%|████▌     | 64/140 [00:06<00:07,  9.53it/s]

 47%|████▋     | 66/140 [00:06<00:07,  9.50it/s]

 48%|████▊     | 67/140 [00:06<00:07,  9.46it/s]

 49%|████▉     | 69/140 [00:07<00:07,  9.69it/s]

 50%|█████     | 70/140 [00:07<00:07,  9.74it/s]

 51%|█████     | 71/140 [00:07<00:07,  9.76it/s]

 51%|█████▏    | 72/140 [00:07<00:06,  9.79it/s]

 52%|█████▏    | 73/140 [00:07<00:06,  9.83it/s]

 53%|█████▎    | 74/140 [00:07<00:06,  9.86it/s]

 54%|█████▎    | 75/140 [00:07<00:06,  9.87it/s]

 54%|█████▍    | 76/140 [00:07<00:06,  9.84it/s]

 55%|█████▌    | 77/140 [00:07<00:06,  9.58it/s]

 56%|█████▌    | 78/140 [00:08<00:06,  9.20it/s]

 56%|█████▋    | 79/140 [00:08<00:06,  9.35it/s]

 58%|█████▊    | 81/140 [00:08<00:06,  9.29it/s]

 59%|█████▊    | 82/140 [00:08<00:06,  9.21it/s]

 59%|█████▉    | 83/140 [00:08<00:06,  9.11it/s]

 60%|██████    | 84/140 [00:08<00:06,  9.13it/s]

 61%|██████    | 85/140 [00:08<00:05,  9.32it/s]

 61%|██████▏   | 86/140 [00:08<00:05,  9.21it/s]

 62%|██████▏   | 87/140 [00:09<00:05,  9.18it/s]

 63%|██████▎   | 88/140 [00:09<00:05,  9.08it/s]

 64%|██████▎   | 89/140 [00:09<00:05,  9.10it/s]

 64%|██████▍   | 90/140 [00:09<00:05,  9.00it/s]

 65%|██████▌   | 91/140 [00:09<00:05,  9.09it/s]

 66%|██████▌   | 92/140 [00:09<00:05,  9.30it/s]

 66%|██████▋   | 93/140 [00:09<00:04,  9.46it/s]

 67%|██████▋   | 94/140 [00:09<00:04,  9.29it/s]

 68%|██████▊   | 95/140 [00:09<00:04,  9.03it/s]

 69%|██████▊   | 96/140 [00:10<00:04,  9.13it/s]

 70%|███████   | 98/140 [00:10<00:04,  9.29it/s]

 71%|███████   | 99/140 [00:10<00:04,  9.32it/s]

 71%|███████▏  | 100/140 [00:10<00:04,  9.46it/s]

 72%|███████▏  | 101/140 [00:10<00:04,  9.29it/s]

 73%|███████▎  | 102/140 [00:10<00:04,  9.35it/s]

 74%|███████▎  | 103/140 [00:10<00:03,  9.51it/s]

 75%|███████▌  | 105/140 [00:10<00:03,  9.77it/s]

 76%|███████▋  | 107/140 [00:11<00:03,  9.63it/s]

 77%|███████▋  | 108/140 [00:11<00:03,  9.52it/s]

 78%|███████▊  | 109/140 [00:11<00:03,  9.61it/s]

 79%|███████▊  | 110/140 [00:11<00:03,  9.63it/s]

 79%|███████▉  | 111/140 [00:11<00:03,  9.66it/s]

 81%|████████  | 113/140 [00:11<00:02,  9.84it/s]

 81%|████████▏ | 114/140 [00:11<00:02,  9.83it/s]

 82%|████████▏ | 115/140 [00:12<00:02,  9.53it/s]

 83%|████████▎ | 116/140 [00:12<00:02,  9.42it/s]

 84%|████████▎ | 117/140 [00:12<00:02,  9.54it/s]

 84%|████████▍ | 118/140 [00:12<00:02,  9.32it/s]

 85%|████████▌ | 119/140 [00:12<00:02,  9.29it/s]

 86%|████████▋ | 121/140 [00:12<00:02,  9.42it/s]

 87%|████████▋ | 122/140 [00:12<00:01,  9.12it/s]

 88%|████████▊ | 123/140 [00:12<00:01,  9.09it/s]

 89%|████████▉ | 125/140 [00:13<00:01,  9.25it/s]

 90%|█████████ | 126/140 [00:13<00:01,  9.23it/s]

 91%|█████████ | 127/140 [00:13<00:01,  9.38it/s]

 91%|█████████▏| 128/140 [00:13<00:01,  9.28it/s]

 92%|█████████▏| 129/140 [00:13<00:01,  9.06it/s]

 93%|█████████▎| 130/140 [00:13<00:01,  8.83it/s]

 94%|█████████▎| 131/140 [00:13<00:01,  8.58it/s]

 94%|█████████▍| 132/140 [00:13<00:00,  8.70it/s]

 95%|█████████▌| 133/140 [00:13<00:00,  9.02it/s]

 96%|█████████▋| 135/140 [00:14<00:00,  9.50it/s]

 98%|█████████▊| 137/140 [00:14<00:00,  9.66it/s]

 99%|█████████▊| 138/140 [00:14<00:00,  9.52it/s]

 99%|█████████▉| 139/140 [00:14<00:00,  9.50it/s]

100%|██████████| 140/140 [00:14<00:00,  8.21it/s]
                                                     

                                                 
{'eval_loss': 0.8963637351989746, 'eval_runtime': 15.1217, 'eval_samples_per_second': 18.516, 'eval_steps_per_second': 9.258, 'epoch': 2.93}

 98%|█████████▊| 2500/2562 [2:07:45<03:01,  2.93s/it]

100%|██████████| 140/140 [00:14<00:00,  8.21it/s]

                                                 /opt/conda/lib/python3.11/site-packages/peft/utils/save_and_load.py:372: UserWarning: Could not find a config file in  - will assume that the vocabulary was not modified.
  warnings.warn(

 98%|█████████▊| 2501/2562 [2:07:50<08:00,  7.88s/it]
 98%|█████████▊| 2502/2562 [2:07:53<06:23,  6.39s/it]
 98%|█████████▊| 2503/2562 [2:07:55<05:14,  5.33s/it]
 98%|█████████▊| 2504/2562 [2:07:58<04:26,  4.60s/it]
 98%|█████████▊| 2505/2562 [2:08:01<03:51,  4.07s/it]
 98%|█████████▊| 2506/2562 [2:08:04<03:27,  3.71s/it]
 98%|█████████▊| 2507/2562 [2:08:07<03:13,  3.52s/it]
 98%|█████████▊| 2508/2562 [2:08:10<03:02,  3.38s/it]
 98%|█████████▊| 2509/2562 [2:08:13<02:49,  3.20s/it]
 98%|█████████▊| 2510/2562 [2:08:16<02:45,  3.18s/it]
                                                     
{'loss': 0.7934, 'grad_norm': 4.920419216156006, 'learning_rate': 2.2533520505211294e-08, 'epoch': 2.94}

 98%|█████████▊| 2510/2562 [2:08:16<02:45,  3.18s/it]
 98%|█████████▊| 2511/2562 [2:08:19<02:39,  3.14s/it]
 98%|█████████▊| 2512/2562 [2:08:22<02:31,  3.04s/it]
 98%|█████████▊| 2513/2562 [2:08:25<02:28,  3.02s/it]
 98%|█████████▊| 2514/2562 [2:08:28<02:22,  2.96s/it]
 98%|█████████▊| 2515/2562 [2:08:31<02:21,  3.00s/it]
 98%|█████████▊| 2516/2562 [2:08:34<02:19,  3.02s/it]
 98%|█████████▊| 2517/2562 [2:08:37<02:14,  2.99s/it]
 98%|█████████▊| 2518/2562 [2:08:40<02:09,  2.95s/it]
 98%|█████████▊| 2519/2562 [2:08:43<02:07,  2.97s/it]
 98%|█████████▊| 2520/2562 [2:08:46<02:03,  2.93s/it]
                                                     
{'loss': 0.7718, 'grad_norm': 4.6601152420043945, 'learning_rate': 1.470204226908134e-08, 'epoch': 2.95}

 98%|█████████▊| 2520/2562 [2:08:46<02:03,  2.93s/it]
 98%|█████████▊| 2521/2562 [2:08:48<01:59,  2.90s/it]
 98%|█████████▊| 2522/2562 [2:08:51<01:57,  2.94s/it]
 98%|█████████▊| 2523/2562 [2:08:54<01:53,  2.90s/it]
 99%|█████████▊| 2524/2562 [2:08:57<01:51,  2.92s/it]
 99%|█████████▊| 2525/2562 [2:09:00<01:47,  2.91s/it]
 99%|█████████▊| 2526/2562 [2:09:03<01:43,  2.88s/it]
 99%|█████████▊| 2527/2562 [2:09:06<01:42,  2.94s/it]
 99%|█████████▊| 2528/2562 [2:09:09<01:38,  2.90s/it]
 99%|█████████▊| 2529/2562 [2:09:12<01:35,  2.90s/it]
 99%|█████████▉| 2530/2562 [2:09:15<01:33,  2.93s/it]
                                                     
{'loss': 0.7794, 'grad_norm': 4.90738582611084, 'learning_rate': 8.535396443124511e-09, 'epoch': 2.96}

 99%|█████████▉| 2530/2562 [2:09:15<01:33,  2.93s/it]
 99%|█████████▉| 2531/2562 [2:09:17<01:29,  2.88s/it]
 99%|█████████▉| 2532/2562 [2:09:20<01:26,  2.87s/it]
 99%|█████████▉| 2533/2562 [2:09:23<01:24,  2.92s/it]
 99%|█████████▉| 2534/2562 [2:09:26<01:21,  2.91s/it]
 99%|█████████▉| 2535/2562 [2:09:29<01:17,  2.85s/it]
 99%|█████████▉| 2536/2562 [2:09:32<01:14,  2.88s/it]
 99%|█████████▉| 2537/2562 [2:09:35<01:13,  2.94s/it]
 99%|█████████▉| 2538/2562 [2:09:38<01:10,  2.92s/it]
 99%|█████████▉| 2539/2562 [2:09:41<01:07,  2.94s/it]
 99%|█████████▉| 2540/2562 [2:09:44<01:04,  2.94s/it]
                                                     
{'loss': 0.7944, 'grad_norm': 4.550679683685303, 'learning_rate': 4.034611182121007e-09, 'epoch': 2.97}

 99%|█████████▉| 2540/2562 [2:09:44<01:04,  2.94s/it]
 99%|█████████▉| 2541/2562 [2:09:47<01:01,  2.94s/it]
 99%|█████████▉| 2542/2562 [2:09:50<00:59,  2.98s/it]
 99%|█████████▉| 2543/2562 [2:09:53<00:56,  2.98s/it]
 99%|█████████▉| 2544/2562 [2:09:56<00:53,  2.96s/it]
 99%|█████████▉| 2545/2562 [2:09:59<00:50,  2.97s/it]
 99%|█████████▉| 2546/2562 [2:10:02<00:47,  2.98s/it]
 99%|█████████▉| 2547/2562 [2:10:04<00:43,  2.93s/it]
 99%|█████████▉| 2548/2562 [2:10:07<00:41,  2.96s/it]
 99%|█████████▉| 2549/2562 [2:10:10<00:37,  2.91s/it]
100%|█████████▉| 2550/2562 [2:10:13<00:34,  2.90s/it]
                                                     
{'loss': 0.7852, 'grad_norm': 5.020516395568848, 'learning_rate': 1.2004368946427758e-09, 'epoch': 2.99}

100%|█████████▉| 2550/2562 [2:10:13<00:34,  2.90s/it]
100%|█████████▉| 2551/2562 [2:10:16<00:32,  2.94s/it]
100%|█████████▉| 2552/2562 [2:10:19<00:29,  2.91s/it]
100%|█████████▉| 2553/2562 [2:10:22<00:25,  2.88s/it]
100%|█████████▉| 2554/2562 [2:10:25<00:23,  2.92s/it]
100%|█████████▉| 2555/2562 [2:10:28<00:20,  2.89s/it]
100%|█████████▉| 2556/2562 [2:10:31<00:17,  2.94s/it]
100%|█████████▉| 2557/2562 [2:10:34<00:14,  2.93s/it]
100%|█████████▉| 2558/2562 [2:10:36<00:11,  2.89s/it]
100%|█████████▉| 2559/2562 [2:10:39<00:08,  2.94s/it]
100%|█████████▉| 2560/2562 [2:10:42<00:05,  2.90s/it]
                                                     
{'loss': 0.7744, 'grad_norm': 4.78999137878418, 'learning_rate': 3.334611793692766e-11, 'epoch': 3.0}

100%|█████████▉| 2560/2562 [2:10:42<00:05,  2.90s/it]
100%|█████████▉| 2561/2562 [2:10:45<00:02,  2.89s/it]
100%|██████████| 2562/2562 [2:10:48<00:00,  2.94s/it]/opt/conda/lib/python3.11/site-packages/peft/utils/save_and_load.py:372: UserWarning: Could not find a config file in  - will assume that the vocabulary was not modified.
  warnings.warn(

                                                     
{'train_runtime': 7849.8598, 'train_samples_per_second': 5.227, 'train_steps_per_second': 0.326, 'train_loss': 1.2505455842528095, 'epoch': 3.0}

100%|██████████| 2562/2562 [2:10:49<00:00,  2.94s/it]
100%|██████████| 2562/2562 [2:10:49<00:00,  3.06s/it]

Saving LoRA checkpoint...
/opt/conda/lib/python3.11/site-packages/peft/utils/save_and_load.py:372: UserWarning: Could not find a config file in  - will assume that the vocabulary was not modified.
  warnings.warn(

Merging and saving final model...
Merging LoRA weights...
Total keys in output: 495
Saved to /workspace/mistral_3_small_flux2_bf16_finetuned.safetensors (35.6 GB)

============================================================
DONE! Fine-tuned TE saved to: /workspace/mistral_3_small_flux2_bf16_finetuned.safetensors
============================================================