BNB-NOVADAX-TRADE / training.log
Alisson990's picture
Upload training.log with huggingface_hub
0dceefa verified
[19:47:48] === PHASE 1 SUPERVISED PRE-TRAINING START ===
[19:47:48] Baixando dataset do HuggingFace...
[19:47:48] (isso pode levar 15-30 min para 10 GB)
[19:52:15] Dataset em: /kaggle/working/dataset
[19:52:15] Binance 1s: 2304 arquivos
[19:52:15] Binance 1m: 0 arquivos (fallback)
[19:52:15] NovaDax trades: 2304 arquivos
[19:52:15] NovaDax klines: 1 arquivos
[19:52:15] NovaDax indexado: 2304 datas de trades, 1 de klines
[19:52:15] Split: train=1461 | val=366 | test=477 dias
[19:52:15] Testando pipeline de dados com o primeiro arquivo...
[19:52:16] OK: 86400 rows | colunas: ['ts_ms', 'bin_ret', 'bin_ret_10', 'bin_ret_30', 'vol_z', 'trades_z']...
[19:52:16] y30 dist: flat=68476 up=8737 dn=9187
[19:52:16] Feats: mean=0.0361 | std=0.3507
[19:52:17] DataLoaders criados
[19:52:17] Modelo: 4,906,507 params (4.91M) | trainable: 4,906,507
[19:52:17] DataParallel em 2 GPUs
[19:52:17] Steps estimados: 4,400 | Steps/época: 24,654
[19:52:24] Tentando baixar checkpoint do HuggingFace...
[19:52:24] Nenhum checkpoint encontrado, iniciando do zero.
[19:52:24] Iniciando treino | step=0 | best_val=inf
[19:52:24] Máximo: 11h | Checkpoint a cada 1h
[19:52:24] === Época 1 | step=0 | elapsed=0.00h ===
[19:52:46] step= 100 | loss=2.5517 | acc30=0.184 | lr=2.61e-04 | 0.01h
[19:53:03] step= 200 | loss=2.1935 | acc30=0.260 | lr=3.00e-04 | 0.01h
[19:53:20] step= 300 | loss=2.3798 | acc30=0.439 | lr=2.99e-04 | 0.02h
[19:53:37] step= 400 | loss=2.7090 | acc30=0.989 | lr=2.97e-04 | 0.02h
[19:53:54] step= 500 | loss=2.8053 | acc30=0.583 | lr=2.95e-04 | 0.03h
[19:54:12] step= 600 | loss=2.5192 | acc30=0.263 | lr=2.91e-04 | 0.03h
[19:54:31] step= 700 | loss=2.2992 | acc30=0.212 | lr=2.87e-04 | 0.04h
[19:54:48] step= 800 | loss=2.2822 | acc30=0.213 | lr=2.82e-04 | 0.04h
[19:55:06] step= 900 | loss=2.3126 | acc30=0.226 | lr=2.77e-04 | 0.04h
[19:55:23] step= 1000 | loss=2.6052 | acc30=0.505 | lr=2.70e-04 | 0.05h
[19:55:23] Validando...
[19:55:40] VAL: loss=2.6031 acc30=0.601 acc60=0.277 acc120=0.342
[19:55:58] step= 1100 | loss=2.7995 | acc30=0.363 | lr=2.63e-04 | 0.06h
[19:56:15] step= 1200 | loss=2.9147 | acc30=0.927 | lr=2.56e-04 | 0.06h
[19:56:33] step= 1300 | loss=2.4326 | acc30=0.359 | lr=2.48e-04 | 0.07h
[19:56:51] step= 1400 | loss=2.3332 | acc30=0.215 | lr=2.39e-04 | 0.07h
[19:57:09] step= 1500 | loss=2.4511 | acc30=0.167 | lr=2.30e-04 | 0.08h
[19:57:26] step= 1600 | loss=2.4280 | acc30=0.173 | lr=2.21e-04 | 0.08h
[19:57:44] step= 1700 | loss=2.3409 | acc30=0.185 | lr=2.11e-04 | 0.09h
[19:58:02] step= 1800 | loss=2.3322 | acc30=0.207 | lr=2.00e-04 | 0.09h
[19:58:19] step= 1900 | loss=2.1135 | acc30=0.278 | lr=1.90e-04 | 0.10h
[19:58:36] step= 2000 | loss=2.1493 | acc30=0.271 | lr=1.79e-04 | 0.10h
[19:58:36] Validando...
[19:58:53] VAL: loss=2.3881 acc30=0.202 acc60=0.277 acc120=0.342
[19:59:11] step= 2100 | loss=2.1894 | acc30=0.257 | lr=1.68e-04 | 0.11h
[19:59:29] step= 2200 | loss=2.2146 | acc30=0.248 | lr=1.57e-04 | 0.12h
[19:59:47] step= 2300 | loss=2.3106 | acc30=0.216 | lr=1.46e-04 | 0.12h
[20:00:04] step= 2400 | loss=2.1249 | acc30=0.286 | lr=1.35e-04 | 0.13h
[20:00:22] step= 2500 | loss=2.4719 | acc30=0.178 | lr=1.24e-04 | 0.13h
[20:00:39] step= 2600 | loss=2.3971 | acc30=0.191 | lr=1.13e-04 | 0.14h
[20:00:57] step= 2700 | loss=2.2963 | acc30=0.210 | lr=1.03e-04 | 0.14h
[20:01:14] step= 2800 | loss=2.6621 | acc30=0.121 | lr=9.25e-05 | 0.15h
[20:01:32] step= 2900 | loss=2.8470 | acc30=0.359 | lr=8.24e-05 | 0.15h
[20:01:50] step= 3000 | loss=2.5497 | acc30=0.147 | lr=7.28e-05 | 0.16h
[20:01:50] Validando...
[20:02:06] VAL: loss=2.3954 acc30=0.197 acc60=0.268 acc120=0.324
[20:02:25] step= 3100 | loss=2.3512 | acc30=0.239 | lr=6.35e-05 | 0.17h
[20:02:42] step= 3200 | loss=2.0462 | acc30=0.324 | lr=5.47e-05 | 0.17h
[20:03:00] step= 3300 | loss=3.5350 | acc30=0.198 | lr=4.65e-05 | 0.18h
[20:03:18] step= 3400 | loss=2.7692 | acc30=0.233 | lr=3.88e-05 | 0.18h
[20:03:36] step= 3500 | loss=2.2814 | acc30=0.249 | lr=3.17e-05 | 0.19h
[20:03:53] step= 3600 | loss=2.8433 | acc30=0.103 | lr=2.52e-05 | 0.19h
[20:04:11] step= 3700 | loss=2.3994 | acc30=0.195 | lr=1.94e-05 | 0.20h
[20:04:28] step= 3800 | loss=2.4279 | acc30=0.175 | lr=1.44e-05 | 0.20h
[20:04:46] step= 3900 | loss=2.5048 | acc30=0.157 | lr=1.00e-05 | 0.21h
[20:05:04] step= 4000 | loss=2.4662 | acc30=0.160 | lr=6.43e-06 | 0.21h
[20:05:04] Validando...
[20:05:20] VAL: loss=2.3599 acc30=0.200 acc60=0.273 acc120=0.337
[20:05:38] step= 4100 | loss=2.3305 | acc30=0.229 | lr=3.63e-06 | 0.22h
[20:05:56] step= 4200 | loss=2.3756 | acc30=0.214 | lr=1.62e-06 | 0.23h
[20:06:14] step= 4300 | loss=2.6299 | acc30=0.112 | lr=4.08e-07 | 0.23h
[20:06:31] step= 4400 | loss=3.2240 | acc30=0.053 | lr=1.00e-08 | 0.24h
[20:06:49] step= 4500 | loss=2.4437 | acc30=0.290 | lr=1.00e-08 | 0.24h
[20:07:07] step= 4600 | loss=2.1073 | acc30=0.389 | lr=1.00e-08 | 0.25h
[20:07:24] step= 4700 | loss=2.2409 | acc30=0.248 | lr=1.00e-08 | 0.25h
[20:07:42] step= 4800 | loss=2.0776 | acc30=0.385 | lr=1.00e-08 | 0.26h
[20:08:00] step= 4900 | loss=2.3494 | acc30=0.258 | lr=1.00e-08 | 0.26h
[20:08:18] step= 5000 | loss=2.5491 | acc30=0.167 | lr=1.00e-08 | 0.27h
[20:08:18] Validando...
[20:08:34] VAL: loss=2.3569 acc30=0.200 acc60=0.272 acc120=0.337
[20:08:52] step= 5100 | loss=2.8098 | acc30=0.088 | lr=1.00e-08 | 0.27h
[20:09:10] step= 5200 | loss=2.3844 | acc30=0.192 | lr=1.00e-08 | 0.28h
[20:09:28] step= 5300 | loss=2.4574 | acc30=0.163 | lr=1.00e-08 | 0.28h
[20:09:45] step= 5400 | loss=2.4677 | acc30=0.173 | lr=1.00e-08 | 0.29h
[20:10:03] step= 5500 | loss=2.1160 | acc30=0.358 | lr=1.00e-08 | 0.29h
[20:10:21] step= 5600 | loss=2.2874 | acc30=0.226 | lr=1.00e-08 | 0.30h
[20:10:39] step= 5700 | loss=2.2475 | acc30=0.276 | lr=1.00e-08 | 0.30h
[20:10:57] step= 5800 | loss=2.4170 | acc30=0.293 | lr=1.00e-08 | 0.31h
[20:11:14] step= 5900 | loss=4.3855 | acc30=0.010 | lr=1.00e-08 | 0.31h
[20:11:32] step= 6000 | loss=3.4007 | acc30=0.065 | lr=1.00e-08 | 0.32h
[20:11:32] Validando...
[20:11:48] VAL: loss=2.3568 acc30=0.200 acc60=0.272 acc120=0.337
[20:12:06] step= 6100 | loss=2.9336 | acc30=0.081 | lr=1.00e-08 | 0.33h
[20:12:24] step= 6200 | loss=2.4307 | acc30=0.189 | lr=1.00e-08 | 0.33h
[20:12:41] step= 6300 | loss=3.0773 | acc30=0.060 | lr=1.00e-08 | 0.34h
[20:12:59] step= 6400 | loss=2.3944 | acc30=0.218 | lr=1.00e-08 | 0.34h
[20:13:17] step= 6500 | loss=2.7768 | acc30=0.123 | lr=1.00e-08 | 0.35h
[20:13:35] step= 6600 | loss=3.0216 | acc30=0.074 | lr=1.00e-08 | 0.35h
[20:13:52] step= 6700 | loss=2.3582 | acc30=0.162 | lr=1.00e-08 | 0.36h
[20:14:10] step= 6800 | loss=2.7971 | acc30=0.096 | lr=1.00e-08 | 0.36h
[20:14:28] step= 6900 | loss=2.4053 | acc30=0.172 | lr=1.00e-08 | 0.37h
[20:14:47] step= 7000 | loss=2.5172 | acc30=0.146 | lr=1.00e-08 | 0.37h
[20:14:47] Validando...
[20:15:03] VAL: loss=2.3568 acc30=0.200 acc60=0.272 acc120=0.337
[20:15:20] step= 7100 | loss=2.2227 | acc30=0.267 | lr=1.00e-08 | 0.38h
[20:15:38] step= 7200 | loss=2.1562 | acc30=0.295 | lr=1.00e-08 | 0.39h
[20:15:56] step= 7300 | loss=2.4936 | acc30=0.211 | lr=1.00e-08 | 0.39h
[20:16:14] step= 7400 | loss=2.2936 | acc30=0.218 | lr=1.00e-08 | 0.40h
[20:16:31] step= 7500 | loss=2.8560 | acc30=0.088 | lr=1.00e-08 | 0.40h
[20:16:49] step= 7600 | loss=2.4172 | acc30=0.177 | lr=1.00e-08 | 0.41h
[20:17:07] step= 7700 | loss=2.3952 | acc30=0.214 | lr=1.00e-08 | 0.41h
[20:17:25] step= 7800 | loss=2.7299 | acc30=0.109 | lr=1.00e-08 | 0.42h
[20:17:42] step= 7900 | loss=2.3964 | acc30=0.184 | lr=1.00e-08 | 0.42h
[20:18:00] step= 8000 | loss=3.2202 | acc30=0.096 | lr=1.00e-08 | 0.43h
[20:18:00] Validando...
[20:18:16] VAL: loss=2.3568 acc30=0.200 acc60=0.272 acc120=0.337
[20:18:34] step= 8100 | loss=2.9130 | acc30=0.117 | lr=1.00e-08 | 0.44h
[20:18:51] step= 8200 | loss=3.6459 | acc30=0.049 | lr=1.00e-08 | 0.44h
[20:19:09] step= 8300 | loss=2.1714 | acc30=0.289 | lr=1.00e-08 | 0.45h
[20:19:27] step= 8400 | loss=2.1737 | acc30=0.303 | lr=1.00e-08 | 0.45h
[20:19:45] step= 8500 | loss=2.3622 | acc30=0.230 | lr=1.00e-08 | 0.46h
[20:20:02] step= 8600 | loss=3.8097 | acc30=0.022 | lr=1.00e-08 | 0.46h
[20:20:20] step= 8700 | loss=4.7551 | acc30=0.011 | lr=1.00e-08 | 0.47h
[20:20:38] step= 8800 | loss=2.5744 | acc30=0.135 | lr=1.00e-08 | 0.47h
[20:20:56] step= 8900 | loss=2.2570 | acc30=0.266 | lr=1.00e-08 | 0.48h
[20:21:13] step= 9000 | loss=2.0862 | acc30=0.397 | lr=1.00e-08 | 0.48h
[20:21:13] Validando...
[20:21:31] VAL: loss=2.3567 acc30=0.200 acc60=0.272 acc120=0.337
[20:21:48] step= 9100 | loss=2.8084 | acc30=0.129 | lr=1.00e-08 | 0.49h
[20:22:06] step= 9200 | loss=2.4362 | acc30=0.148 | lr=1.00e-08 | 0.50h
[20:22:24] step= 9300 | loss=2.2917 | acc30=0.217 | lr=1.00e-08 | 0.50h
[20:22:41] step= 9400 | loss=2.0908 | acc30=0.378 | lr=1.00e-08 | 0.50h
[20:22:59] step= 9500 | loss=2.1227 | acc30=0.321 | lr=1.00e-08 | 0.51h
[20:23:17] step= 9600 | loss=2.8909 | acc30=0.176 | lr=1.00e-08 | 0.51h
[20:23:35] step= 9700 | loss=2.9897 | acc30=0.115 | lr=1.00e-08 | 0.52h
[20:23:52] step= 9800 | loss=2.1571 | acc30=0.311 | lr=1.00e-08 | 0.52h
[20:24:10] step= 9900 | loss=2.2028 | acc30=0.266 | lr=1.00e-08 | 0.53h
[20:24:28] step= 10000 | loss=2.2150 | acc30=0.270 | lr=1.00e-08 | 0.53h
[20:24:28] Validando...
[20:24:44] VAL: loss=2.3566 acc30=0.200 acc60=0.272 acc120=0.337
[20:25:02] step= 10100 | loss=2.2938 | acc30=0.205 | lr=1.00e-08 | 0.54h
[20:25:19] step= 10200 | loss=2.3379 | acc30=0.220 | lr=1.00e-08 | 0.55h
[20:25:37] step= 10300 | loss=2.2540 | acc30=0.234 | lr=1.00e-08 | 0.55h
[20:25:55] step= 10400 | loss=2.1680 | acc30=0.299 | lr=1.00e-08 | 0.56h
[20:26:13] step= 10500 | loss=2.1737 | acc30=0.288 | lr=1.00e-08 | 0.56h
[20:26:30] step= 10600 | loss=2.2804 | acc30=0.209 | lr=1.00e-08 | 0.57h
[20:26:48] step= 10700 | loss=2.2547 | acc30=0.218 | lr=1.00e-08 | 0.57h
[20:27:06] step= 10800 | loss=2.7201 | acc30=0.160 | lr=1.00e-08 | 0.58h
[20:27:24] step= 10900 | loss=3.1726 | acc30=0.065 | lr=1.00e-08 | 0.58h
[20:27:41] step= 11000 | loss=3.3902 | acc30=0.045 | lr=1.00e-08 | 0.59h
[20:27:41] Validando...
[20:27:57] VAL: loss=2.3565 acc30=0.200 acc60=0.272 acc120=0.337
[20:28:15] step= 11100 | loss=2.7160 | acc30=0.150 | lr=1.00e-08 | 0.60h
[20:28:33] step= 11200 | loss=2.3661 | acc30=0.193 | lr=1.00e-08 | 0.60h
[20:28:51] step= 11300 | loss=2.5008 | acc30=0.194 | lr=1.00e-08 | 0.61h
[20:29:09] step= 11400 | loss=3.8704 | acc30=0.027 | lr=1.00e-08 | 0.61h
[20:29:26] step= 11500 | loss=2.7116 | acc30=0.140 | lr=1.00e-08 | 0.62h
[20:29:44] step= 11600 | loss=2.2880 | acc30=0.243 | lr=1.00e-08 | 0.62h
[20:30:02] step= 11700 | loss=2.1555 | acc30=0.309 | lr=1.00e-08 | 0.63h
[20:30:19] step= 11800 | loss=2.2513 | acc30=0.252 | lr=1.00e-08 | 0.63h
[20:30:37] step= 11900 | loss=2.1663 | acc30=0.290 | lr=1.00e-08 | 0.64h
[20:30:55] step= 12000 | loss=2.1389 | acc30=0.330 | lr=1.00e-08 | 0.64h
[20:30:55] Validando...
[20:31:11] VAL: loss=2.3565 acc30=0.200 acc60=0.272 acc120=0.337
[20:31:29] step= 12100 | loss=2.2540 | acc30=0.261 | lr=1.00e-08 | 0.65h
[20:31:46] step= 12200 | loss=3.5529 | acc30=0.050 | lr=1.00e-08 | 0.66h
[20:32:04] step= 12300 | loss=2.2384 | acc30=0.274 | lr=1.00e-08 | 0.66h
[20:32:22] step= 12400 | loss=2.1568 | acc30=0.307 | lr=1.00e-08 | 0.67h
[20:32:40] step= 12500 | loss=2.0656 | acc30=0.405 | lr=1.00e-08 | 0.67h
[20:32:57] step= 12600 | loss=2.2727 | acc30=0.237 | lr=1.00e-08 | 0.68h
[20:33:15] step= 12700 | loss=2.2428 | acc30=0.243 | lr=1.00e-08 | 0.68h
[20:33:33] step= 12800 | loss=2.5467 | acc30=0.206 | lr=1.00e-08 | 0.69h
[20:33:51] step= 12900 | loss=3.3666 | acc30=0.058 | lr=1.00e-08 | 0.69h
[20:34:08] step= 13000 | loss=2.1594 | acc30=0.297 | lr=1.00e-08 | 0.70h
[20:34:08] Validando...
[20:34:25] VAL: loss=2.3564 acc30=0.200 acc60=0.272 acc120=0.337
[20:34:43] step= 13100 | loss=2.1094 | acc30=0.344 | lr=1.00e-08 | 0.71h
[20:35:01] step= 13200 | loss=2.2251 | acc30=0.279 | lr=1.00e-08 | 0.71h
[20:35:18] step= 13300 | loss=2.3688 | acc30=0.197 | lr=1.00e-08 | 0.72h
[20:35:36] step= 13400 | loss=2.2561 | acc30=0.242 | lr=1.00e-08 | 0.72h
[20:35:54] step= 13500 | loss=2.3765 | acc30=0.180 | lr=1.00e-08 | 0.73h
[20:36:12] step= 13600 | loss=2.4212 | acc30=0.171 | lr=1.00e-08 | 0.73h
[20:36:29] step= 13700 | loss=2.4924 | acc30=0.158 | lr=1.00e-08 | 0.73h
[20:36:47] step= 13800 | loss=2.3785 | acc30=0.198 | lr=1.00e-08 | 0.74h
[20:37:05] step= 13900 | loss=2.3181 | acc30=0.195 | lr=1.00e-08 | 0.74h
[20:37:23] step= 14000 | loss=2.1844 | acc30=0.263 | lr=1.00e-08 | 0.75h
[20:37:23] Validando...
[20:37:39] VAL: loss=2.3563 acc30=0.200 acc60=0.271 acc120=0.337
[20:37:57] step= 14100 | loss=2.2146 | acc30=0.258 | lr=1.00e-08 | 0.76h
[20:38:15] step= 14200 | loss=2.0913 | acc30=0.388 | lr=1.00e-08 | 0.76h
[20:38:32] step= 14300 | loss=3.8682 | acc30=0.156 | lr=1.00e-08 | 0.77h
[20:38:50] step= 14400 | loss=4.0869 | acc30=0.049 | lr=1.00e-08 | 0.77h
[20:39:07] step= 14500 | loss=2.3883 | acc30=0.184 | lr=1.00e-08 | 0.78h
[20:39:25] step= 14600 | loss=2.6680 | acc30=0.115 | lr=1.00e-08 | 0.78h
[20:39:43] step= 14700 | loss=2.5179 | acc30=0.137 | lr=1.00e-08 | 0.79h
[20:40:01] step= 14800 | loss=2.4570 | acc30=0.157 | lr=1.00e-08 | 0.79h
[20:40:18] step= 14900 | loss=2.5993 | acc30=0.135 | lr=1.00e-08 | 0.80h
[20:40:36] step= 15000 | loss=2.3919 | acc30=0.185 | lr=1.00e-08 | 0.80h
[20:40:36] Validando...
[20:40:53] VAL: loss=2.3562 acc30=0.200 acc60=0.272 acc120=0.337
[20:41:11] step= 15100 | loss=2.1487 | acc30=0.298 | lr=1.00e-08 | 0.81h
[20:41:29] step= 15200 | loss=2.1501 | acc30=0.312 | lr=1.00e-08 | 0.82h
[20:41:46] step= 15300 | loss=2.3261 | acc30=0.211 | lr=1.00e-08 | 0.82h
[20:42:04] step= 15400 | loss=2.1218 | acc30=0.383 | lr=1.00e-08 | 0.83h
[20:42:22] step= 15500 | loss=2.3623 | acc30=0.280 | lr=1.00e-08 | 0.83h
[20:42:40] step= 15600 | loss=2.5723 | acc30=0.124 | lr=1.00e-08 | 0.84h
[20:42:57] step= 15700 | loss=2.4892 | acc30=0.155 | lr=1.00e-08 | 0.84h
[20:43:15] step= 15800 | loss=2.4106 | acc30=0.187 | lr=1.00e-08 | 0.85h
[20:43:33] step= 15900 | loss=2.4061 | acc30=0.168 | lr=1.00e-08 | 0.85h
[20:43:51] step= 16000 | loss=2.2963 | acc30=0.221 | lr=1.00e-08 | 0.86h
[20:43:51] Validando...
[20:44:07] VAL: loss=2.3562 acc30=0.200 acc60=0.272 acc120=0.337
[20:44:24] step= 16100 | loss=2.1279 | acc30=0.334 | lr=1.00e-08 | 0.87h
[20:44:42] step= 16200 | loss=4.1668 | acc30=0.086 | lr=1.00e-08 | 0.87h
[20:45:00] step= 16300 | loss=3.8789 | acc30=0.062 | lr=1.00e-08 | 0.88h
[20:45:18] step= 16400 | loss=2.2729 | acc30=0.221 | lr=1.00e-08 | 0.88h
[20:45:35] step= 16500 | loss=2.1995 | acc30=0.280 | lr=1.00e-08 | 0.89h
[20:45:53] step= 16600 | loss=2.2381 | acc30=0.237 | lr=1.00e-08 | 0.89h
[20:46:11] step= 16700 | loss=2.1802 | acc30=0.273 | lr=1.00e-08 | 0.90h
[20:46:29] step= 16800 | loss=2.1204 | acc30=0.316 | lr=1.00e-08 | 0.90h
[20:46:46] step= 16900 | loss=2.3939 | acc30=0.181 | lr=1.00e-08 | 0.91h
[20:47:04] step= 17000 | loss=2.2572 | acc30=0.222 | lr=1.00e-08 | 0.91h
[20:47:04] Validando...
[20:47:20] VAL: loss=2.3561 acc30=0.200 acc60=0.272 acc120=0.337
[20:47:38] step= 17100 | loss=2.2856 | acc30=0.219 | lr=1.00e-08 | 0.92h
[20:47:56] step= 17200 | loss=2.4377 | acc30=0.175 | lr=1.00e-08 | 0.93h
[20:48:13] step= 17300 | loss=2.5572 | acc30=0.148 | lr=1.00e-08 | 0.93h
[20:48:31] step= 17400 | loss=2.1858 | acc30=0.307 | lr=1.00e-08 | 0.94h
[20:48:49] step= 17500 | loss=2.3156 | acc30=0.289 | lr=1.00e-08 | 0.94h
[20:49:06] step= 17600 | loss=2.6825 | acc30=0.131 | lr=1.00e-08 | 0.95h
[20:49:24] step= 17700 | loss=2.5440 | acc30=0.152 | lr=1.00e-08 | 0.95h
[20:49:42] step= 17800 | loss=2.2803 | acc30=0.222 | lr=1.00e-08 | 0.96h
[20:50:00] step= 17900 | loss=2.2970 | acc30=0.229 | lr=1.00e-08 | 0.96h
[20:50:17] step= 18000 | loss=2.1674 | acc30=0.288 | lr=1.00e-08 | 0.96h
[20:50:17] Validando...
[20:50:35] VAL: loss=2.3560 acc30=0.200 acc60=0.272 acc120=0.337
[20:50:52] step= 18100 | loss=3.2475 | acc30=0.071 | lr=1.00e-08 | 0.97h
[20:51:10] step= 18200 | loss=2.5619 | acc30=0.143 | lr=1.00e-08 | 0.98h
[20:51:28] step= 18300 | loss=2.1799 | acc30=0.278 | lr=1.00e-08 | 0.98h
[20:51:45] step= 18400 | loss=2.1331 | acc30=0.299 | lr=1.00e-08 | 0.99h
[20:52:03] step= 18500 | loss=2.2125 | acc30=0.269 | lr=1.00e-08 | 0.99h
[20:52:21] step= 18600 | loss=2.1454 | acc30=0.307 | lr=1.00e-08 | 1.00h
[20:52:24] --- CHECKPOINT (hora 1.0) ---
[20:52:24] Checkpoint salvo: checkpoint_step00018616.pt (59.0 MB)
[20:52:30] ☁ Upload HuggingFace OK: step=18616
[20:52:40] step= 18700 | loss=2.2653 | acc30=0.270 | lr=1.00e-08 | 1.00h
[20:52:57] step= 18800 | loss=2.5632 | acc30=0.160 | lr=1.00e-08 | 1.01h
[20:53:15] step= 18900 | loss=2.6585 | acc30=0.120 | lr=1.00e-08 | 1.01h
[20:53:33] step= 19000 | loss=2.3882 | acc30=0.188 | lr=1.00e-08 | 1.02h
[20:53:33] Validando...
[20:53:49] VAL: loss=2.3559 acc30=0.200 acc60=0.271 acc120=0.337
[20:54:07] step= 19100 | loss=2.3161 | acc30=0.217 | lr=1.00e-08 | 1.03h
[20:54:24] step= 19200 | loss=2.2796 | acc30=0.214 | lr=1.00e-08 | 1.03h
[20:54:42] step= 19300 | loss=2.3702 | acc30=0.178 | lr=1.00e-08 | 1.04h
[20:55:00] step= 19400 | loss=2.3311 | acc30=0.197 | lr=1.00e-08 | 1.04h
[20:55:18] step= 19500 | loss=2.2664 | acc30=0.229 | lr=1.00e-08 | 1.05h
[20:55:35] step= 19600 | loss=2.1307 | acc30=0.333 | lr=1.00e-08 | 1.05h
[20:55:53] step= 19700 | loss=2.6205 | acc30=0.166 | lr=1.00e-08 | 1.06h
[20:56:11] step= 19800 | loss=2.2868 | acc30=0.236 | lr=1.00e-08 | 1.06h
[20:56:29] step= 19900 | loss=2.3121 | acc30=0.231 | lr=1.00e-08 | 1.07h
[20:56:46] step= 20000 | loss=2.5178 | acc30=0.147 | lr=1.00e-08 | 1.07h
[20:56:46] Validando...
[20:57:02] VAL: loss=2.3558 acc30=0.200 acc60=0.271 acc120=0.337
[20:57:20] step= 20100 | loss=3.0066 | acc30=0.094 | lr=1.00e-08 | 1.08h
[20:57:38] step= 20200 | loss=2.2709 | acc30=0.261 | lr=1.00e-08 | 1.09h
[20:57:56] step= 20300 | loss=2.1522 | acc30=0.313 | lr=1.00e-08 | 1.09h
[20:58:13] step= 20400 | loss=2.8805 | acc30=0.085 | lr=1.00e-08 | 1.10h
[20:58:31] step= 20500 | loss=2.2893 | acc30=0.236 | lr=1.00e-08 | 1.10h
[20:58:49] step= 20600 | loss=2.2719 | acc30=0.230 | lr=1.00e-08 | 1.11h
[20:59:07] step= 20700 | loss=2.3882 | acc30=0.199 | lr=1.00e-08 | 1.11h
[20:59:24] step= 20800 | loss=3.6072 | acc30=0.082 | lr=1.00e-08 | 1.12h
[20:59:42] step= 20900 | loss=2.4049 | acc30=0.168 | lr=1.00e-08 | 1.12h
[21:00:00] step= 21000 | loss=2.1991 | acc30=0.273 | lr=1.00e-08 | 1.13h
[21:00:00] Validando...
[21:00:16] VAL: loss=2.3557 acc30=0.200 acc60=0.272 acc120=0.337
[21:00:34] step= 21100 | loss=2.1083 | acc30=0.350 | lr=1.00e-08 | 1.14h
[21:00:51] step= 21200 | loss=2.2334 | acc30=0.239 | lr=1.00e-08 | 1.14h
[21:01:09] step= 21300 | loss=2.7824 | acc30=0.117 | lr=1.00e-08 | 1.15h
[21:01:27] step= 21400 | loss=2.4577 | acc30=0.188 | lr=1.00e-08 | 1.15h
[21:01:45] step= 21500 | loss=2.2416 | acc30=0.231 | lr=1.00e-08 | 1.16h
[21:02:02] step= 21600 | loss=2.2314 | acc30=0.246 | lr=1.00e-08 | 1.16h
[21:02:20] step= 21700 | loss=2.8155 | acc30=0.091 | lr=1.00e-08 | 1.17h
[21:02:38] step= 21800 | loss=2.4696 | acc30=0.162 | lr=1.00e-08 | 1.17h
[21:02:55] step= 21900 | loss=2.5120 | acc30=0.131 | lr=1.00e-08 | 1.18h
[21:03:13] step= 22000 | loss=2.6247 | acc30=0.110 | lr=1.00e-08 | 1.18h
[21:03:13] Validando...
[21:03:29] VAL: loss=2.3557 acc30=0.200 acc60=0.272 acc120=0.337
[21:03:47] step= 22100 | loss=2.2786 | acc30=0.274 | lr=1.00e-08 | 1.19h
[21:04:05] step= 22200 | loss=2.1456 | acc30=0.337 | lr=1.00e-08 | 1.19h
[21:04:22] step= 22300 | loss=2.2866 | acc30=0.215 | lr=1.00e-08 | 1.20h
[21:04:40] step= 22400 | loss=2.9401 | acc30=0.074 | lr=1.00e-08 | 1.20h
[21:04:58] step= 22500 | loss=2.6159 | acc30=0.165 | lr=1.00e-08 | 1.21h
[21:05:16] step= 22600 | loss=2.2057 | acc30=0.267 | lr=1.00e-08 | 1.21h
[21:05:34] step= 22700 | loss=2.3951 | acc30=0.173 | lr=1.00e-08 | 1.22h
[21:05:52] step= 22800 | loss=2.5167 | acc30=0.142 | lr=1.00e-08 | 1.22h
[21:06:10] step= 22900 | loss=2.6317 | acc30=0.140 | lr=1.00e-08 | 1.23h
[21:06:28] step= 23000 | loss=2.3489 | acc30=0.232 | lr=1.00e-08 | 1.23h
[21:06:28] Validando...
[21:06:44] VAL: loss=2.3556 acc30=0.200 acc60=0.272 acc120=0.337
[21:07:01] step= 23100 | loss=2.1007 | acc30=0.360 | lr=1.00e-08 | 1.24h
[21:07:19] step= 23200 | loss=3.1069 | acc30=0.105 | lr=1.00e-08 | 1.25h
[21:07:37] step= 23300 | loss=2.5647 | acc30=0.138 | lr=1.00e-08 | 1.25h
[21:07:55] step= 23400 | loss=2.4188 | acc30=0.170 | lr=1.00e-08 | 1.26h
[21:08:12] step= 23500 | loss=2.5625 | acc30=0.161 | lr=1.00e-08 | 1.26h
[21:08:30] step= 23600 | loss=2.9813 | acc30=0.075 | lr=1.00e-08 | 1.27h
[21:08:48] step= 23700 | loss=2.7562 | acc30=0.112 | lr=1.00e-08 | 1.27h
[21:09:06] step= 23800 | loss=2.2523 | acc30=0.239 | lr=1.00e-08 | 1.28h
[21:09:23] step= 23900 | loss=2.1015 | acc30=0.373 | lr=1.00e-08 | 1.28h
[21:09:41] step= 24000 | loss=2.1484 | acc30=0.305 | lr=1.00e-08 | 1.29h
[21:09:41] Validando...
[21:09:57] VAL: loss=2.3556 acc30=0.200 acc60=0.272 acc120=0.337
[21:10:15] step= 24100 | loss=2.4510 | acc30=0.207 | lr=1.00e-08 | 1.30h
[21:10:33] step= 24200 | loss=2.2686 | acc30=0.226 | lr=1.00e-08 | 1.30h
[21:10:51] step= 24300 | loss=2.2630 | acc30=0.232 | lr=1.00e-08 | 1.31h
[21:11:08] step= 24400 | loss=2.4693 | acc30=0.176 | lr=1.00e-08 | 1.31h
[21:11:27] step= 24500 | loss=2.2745 | acc30=0.243 | lr=1.00e-08 | 1.32h
[21:11:44] step= 24600 | loss=2.1392 | acc30=0.319 | lr=1.00e-08 | 1.32h
[21:12:02] step= 24700 | loss=2.2465 | acc30=0.251 | lr=1.00e-08 | 1.33h
[21:12:20] step= 24800 | loss=2.1234 | acc30=0.341 | lr=1.00e-08 | 1.33h
[21:12:38] step= 24900 | loss=2.2430 | acc30=0.302 | lr=1.00e-08 | 1.34h
[21:12:56] step= 25000 | loss=2.3505 | acc30=0.194 | lr=1.00e-08 | 1.34h
[21:12:56] Validando...
[21:13:12] VAL: loss=2.3555 acc30=0.200 acc60=0.272 acc120=0.337
[21:13:29] step= 25100 | loss=2.6079 | acc30=0.131 | lr=1.00e-08 | 1.35h
[21:13:47] step= 25200 | loss=2.3761 | acc30=0.178 | lr=1.00e-08 | 1.36h
[21:14:05] step= 25300 | loss=2.3850 | acc30=0.191 | lr=1.00e-08 | 1.36h
[21:14:22] step= 25400 | loss=2.4960 | acc30=0.169 | lr=1.00e-08 | 1.37h
[21:14:40] step= 25500 | loss=2.6858 | acc30=0.107 | lr=1.00e-08 | 1.37h
[21:14:58] step= 25600 | loss=2.4176 | acc30=0.181 | lr=1.00e-08 | 1.38h
[21:15:16] step= 25700 | loss=2.5903 | acc30=0.194 | lr=1.00e-08 | 1.38h
[21:15:34] step= 25800 | loss=3.7942 | acc30=0.028 | lr=1.00e-08 | 1.39h
[21:15:51] step= 25900 | loss=2.8155 | acc30=0.134 | lr=1.00e-08 | 1.39h
[21:16:09] step= 26000 | loss=2.3299 | acc30=0.214 | lr=1.00e-08 | 1.40h
[21:16:09] Validando...
[21:16:25] VAL: loss=2.3555 acc30=0.200 acc60=0.272 acc120=0.337
[21:16:43] step= 26100 | loss=2.2290 | acc30=0.255 | lr=1.00e-08 | 1.41h
[21:17:01] step= 26200 | loss=2.1071 | acc30=0.364 | lr=1.00e-08 | 1.41h
[21:17:19] step= 26300 | loss=2.1796 | acc30=0.259 | lr=1.00e-08 | 1.42h
[21:17:37] step= 26400 | loss=2.2602 | acc30=0.218 | lr=1.00e-08 | 1.42h
[21:17:55] step= 26500 | loss=2.2570 | acc30=0.228 | lr=1.00e-08 | 1.43h
[21:18:12] step= 26600 | loss=2.3091 | acc30=0.227 | lr=1.00e-08 | 1.43h
[21:18:30] step= 26700 | loss=2.3590 | acc30=0.182 | lr=1.00e-08 | 1.44h
[21:18:48] step= 26800 | loss=2.1955 | acc30=0.263 | lr=1.00e-08 | 1.44h
[21:19:06] step= 26900 | loss=2.2342 | acc30=0.240 | lr=1.00e-08 | 1.45h
[21:19:23] step= 27000 | loss=2.2254 | acc30=0.264 | lr=1.00e-08 | 1.45h
[21:19:23] Validando...
[21:19:39] VAL: loss=2.3553 acc30=0.200 acc60=0.272 acc120=0.337
[21:19:57] step= 27100 | loss=2.1853 | acc30=0.304 | lr=1.00e-08 | 1.46h
[21:20:15] step= 27200 | loss=2.2296 | acc30=0.247 | lr=1.00e-08 | 1.46h
[21:20:33] step= 27300 | loss=2.3044 | acc30=0.212 | lr=1.00e-08 | 1.47h
[21:20:50] step= 27400 | loss=2.2153 | acc30=0.263 | lr=1.00e-08 | 1.47h
[21:21:08] step= 27500 | loss=3.2300 | acc30=0.077 | lr=1.00e-08 | 1.48h
[21:21:26] step= 27600 | loss=2.4271 | acc30=0.218 | lr=1.00e-08 | 1.48h
[21:21:44] step= 27700 | loss=2.2946 | acc30=0.282 | lr=1.00e-08 | 1.49h
[21:22:01] step= 27800 | loss=2.8539 | acc30=0.075 | lr=1.00e-08 | 1.49h
[21:22:19] step= 27900 | loss=3.1361 | acc30=0.061 | lr=1.00e-08 | 1.50h
[21:22:37] step= 28000 | loss=2.5440 | acc30=0.144 | lr=1.00e-08 | 1.50h
[21:22:38] Validando...
[21:22:53] VAL: loss=2.3553 acc30=0.200 acc60=0.272 acc120=0.337
[21:23:11] step= 28100 | loss=2.1303 | acc30=0.324 | lr=1.00e-08 | 1.51h
[21:23:29] step= 28200 | loss=2.2953 | acc30=0.210 | lr=1.00e-08 | 1.52h
[21:23:47] step= 28300 | loss=2.5547 | acc30=0.153 | lr=1.00e-08 | 1.52h
[21:24:05] step= 28400 | loss=2.5972 | acc30=0.146 | lr=1.00e-08 | 1.53h
[21:24:23] step= 28500 | loss=2.2504 | acc30=0.248 | lr=1.00e-08 | 1.53h
[21:24:40] step= 28600 | loss=2.2301 | acc30=0.240 | lr=1.00e-08 | 1.54h
[21:24:58] step= 28700 | loss=2.3213 | acc30=0.210 | lr=1.00e-08 | 1.54h
[21:25:16] step= 28800 | loss=2.1801 | acc30=0.273 | lr=1.00e-08 | 1.55h
[21:25:34] step= 28900 | loss=2.1934 | acc30=0.262 | lr=1.00e-08 | 1.55h
[21:25:51] step= 29000 | loss=2.2187 | acc30=0.254 | lr=1.00e-08 | 1.56h
[21:25:51] Validando...
[21:26:07] VAL: loss=2.3552 acc30=0.200 acc60=0.271 acc120=0.337
[21:26:25] step= 29100 | loss=3.2036 | acc30=0.075 | lr=1.00e-08 | 1.57h
[21:26:43] step= 29200 | loss=2.5857 | acc30=0.183 | lr=1.00e-08 | 1.57h
[21:27:01] step= 29300 | loss=2.1596 | acc30=0.302 | lr=1.00e-08 | 1.58h
[21:27:19] step= 29400 | loss=2.0846 | acc30=0.372 | lr=1.00e-08 | 1.58h
[21:27:36] step= 29500 | loss=2.1283 | acc30=0.312 | lr=1.00e-08 | 1.59h
[21:27:55] step= 29600 | loss=2.3438 | acc30=0.209 | lr=1.00e-08 | 1.59h
[21:28:12] step= 29700 | loss=2.4148 | acc30=0.168 | lr=1.00e-08 | 1.60h
[21:28:30] step= 29800 | loss=2.6532 | acc30=0.136 | lr=1.00e-08 | 1.60h
[21:28:48] step= 29900 | loss=2.4263 | acc30=0.158 | lr=1.00e-08 | 1.61h
[21:29:06] step= 30000 | loss=2.6111 | acc30=0.155 | lr=1.00e-08 | 1.61h
[21:29:06] Validando...
[21:29:22] VAL: loss=2.3552 acc30=0.200 acc60=0.271 acc120=0.337
[21:29:39] step= 30100 | loss=2.6360 | acc30=0.126 | lr=1.00e-08 | 1.62h
[21:29:57] step= 30200 | loss=2.1322 | acc30=0.341 | lr=1.00e-08 | 1.63h
[21:30:15] step= 30300 | loss=2.1723 | acc30=0.308 | lr=1.00e-08 | 1.63h
[21:30:33] step= 30400 | loss=2.1366 | acc30=0.320 | lr=1.00e-08 | 1.64h
[21:30:51] step= 30500 | loss=2.0768 | acc30=0.371 | lr=1.00e-08 | 1.64h
[21:31:09] step= 30600 | loss=2.4922 | acc30=0.163 | lr=1.00e-08 | 1.65h
[21:31:26] step= 30700 | loss=2.4209 | acc30=0.201 | lr=1.00e-08 | 1.65h
[21:31:44] step= 30800 | loss=2.1846 | acc30=0.277 | lr=1.00e-08 | 1.66h
[21:32:02] step= 30900 | loss=2.0826 | acc30=0.365 | lr=1.00e-08 | 1.66h
[21:32:19] step= 31000 | loss=2.2091 | acc30=0.279 | lr=1.00e-08 | 1.67h
[21:32:19] Validando...
[21:32:36] VAL: loss=2.3551 acc30=0.200 acc60=0.272 acc120=0.337
[21:32:54] step= 31100 | loss=2.6067 | acc30=0.166 | lr=1.00e-08 | 1.67h
[21:33:12] step= 31200 | loss=2.7065 | acc30=0.107 | lr=1.00e-08 | 1.68h
[21:33:29] step= 31300 | loss=3.0895 | acc30=0.063 | lr=1.00e-08 | 1.68h
[21:33:47] step= 31400 | loss=2.6038 | acc30=0.121 | lr=1.00e-08 | 1.69h
[21:34:05] step= 31500 | loss=2.9311 | acc30=0.147 | lr=1.00e-08 | 1.69h
[21:34:23] step= 31600 | loss=2.9101 | acc30=0.112 | lr=1.00e-08 | 1.70h
[21:34:40] step= 31700 | loss=2.1630 | acc30=0.280 | lr=1.00e-08 | 1.70h
[21:34:58] step= 31800 | loss=2.2250 | acc30=0.234 | lr=1.00e-08 | 1.71h
[21:35:16] step= 31900 | loss=2.6049 | acc30=0.153 | lr=1.00e-08 | 1.71h
[21:35:34] step= 32000 | loss=2.5042 | acc30=0.173 | lr=1.00e-08 | 1.72h
[21:35:34] Validando...
[21:35:50] VAL: loss=2.3551 acc30=0.200 acc60=0.272 acc120=0.337
[21:36:07] step= 32100 | loss=2.2522 | acc30=0.249 | lr=1.00e-08 | 1.73h
[21:36:25] step= 32200 | loss=2.3553 | acc30=0.194 | lr=1.00e-08 | 1.73h
[21:36:43] step= 32300 | loss=2.2311 | acc30=0.256 | lr=1.00e-08 | 1.74h
[21:37:01] step= 32400 | loss=2.2894 | acc30=0.233 | lr=1.00e-08 | 1.74h
[21:37:19] step= 32500 | loss=2.6332 | acc30=0.118 | lr=1.00e-08 | 1.75h
[21:37:36] step= 32600 | loss=2.8015 | acc30=0.106 | lr=1.00e-08 | 1.75h
[21:37:54] step= 32700 | loss=2.4998 | acc30=0.155 | lr=1.00e-08 | 1.76h
[21:38:13] step= 32800 | loss=2.4615 | acc30=0.146 | lr=1.00e-08 | 1.76h
[21:38:30] step= 32900 | loss=2.3298 | acc30=0.202 | lr=1.00e-08 | 1.77h
[21:38:48] step= 33000 | loss=2.6266 | acc30=0.150 | lr=1.00e-08 | 1.77h
[21:38:48] Validando...
[21:39:04] VAL: loss=2.3550 acc30=0.200 acc60=0.272 acc120=0.337
[21:39:22] step= 33100 | loss=2.3948 | acc30=0.169 | lr=1.00e-08 | 1.78h
[21:39:40] step= 33200 | loss=2.1576 | acc30=0.287 | lr=1.00e-08 | 1.79h
[21:39:57] step= 33300 | loss=2.4288 | acc30=0.178 | lr=1.00e-08 | 1.79h
[21:40:15] step= 33400 | loss=2.4396 | acc30=0.165 | lr=1.00e-08 | 1.80h
[21:40:33] step= 33500 | loss=2.5946 | acc30=0.131 | lr=1.00e-08 | 1.80h
[21:40:50] step= 33600 | loss=2.4976 | acc30=0.180 | lr=1.00e-08 | 1.81h
[21:41:08] step= 33700 | loss=2.5221 | acc30=0.146 | lr=1.00e-08 | 1.81h
[21:41:26] step= 33800 | loss=2.3725 | acc30=0.176 | lr=1.00e-08 | 1.82h
[21:41:44] step= 33900 | loss=2.6732 | acc30=0.144 | lr=1.00e-08 | 1.82h
[21:42:01] step= 34000 | loss=3.1282 | acc30=0.062 | lr=1.00e-08 | 1.83h
[21:42:01] Validando...
[21:42:19] VAL: loss=2.3550 acc30=0.200 acc60=0.272 acc120=0.337
[21:42:36] step= 34100 | loss=2.8960 | acc30=0.077 | lr=1.00e-08 | 1.84h
[21:42:54] step= 34200 | loss=2.7729 | acc30=0.096 | lr=1.00e-08 | 1.84h
[21:43:12] step= 34300 | loss=2.4294 | acc30=0.172 | lr=1.00e-08 | 1.85h
[21:43:29] step= 34400 | loss=2.2144 | acc30=0.255 | lr=1.00e-08 | 1.85h
[21:43:47] step= 34500 | loss=2.2323 | acc30=0.237 | lr=1.00e-08 | 1.86h
[21:44:05] step= 34600 | loss=3.2022 | acc30=0.103 | lr=1.00e-08 | 1.86h
[21:44:23] step= 34700 | loss=2.8659 | acc30=0.128 | lr=1.00e-08 | 1.87h
[21:44:41] step= 34800 | loss=2.2527 | acc30=0.228 | lr=1.00e-08 | 1.87h
[21:44:59] step= 34900 | loss=2.7820 | acc30=0.116 | lr=1.00e-08 | 1.88h
[21:45:17] step= 35000 | loss=2.5744 | acc30=0.202 | lr=1.00e-08 | 1.88h
[21:45:17] Validando...
[21:45:33] VAL: loss=2.3550 acc30=0.200 acc60=0.271 acc120=0.337
[21:45:51] step= 35100 | loss=2.2151 | acc30=0.254 | lr=1.00e-08 | 1.89h
[21:46:08] step= 35200 | loss=2.2439 | acc30=0.228 | lr=1.00e-08 | 1.90h
[21:46:26] step= 35300 | loss=2.1353 | acc30=0.311 | lr=1.00e-08 | 1.90h
[21:46:44] step= 35400 | loss=2.2963 | acc30=0.264 | lr=1.00e-08 | 1.91h
[21:47:02] step= 35500 | loss=2.5508 | acc30=0.144 | lr=1.00e-08 | 1.91h
[21:47:19] step= 35600 | loss=2.6287 | acc30=0.160 | lr=1.00e-08 | 1.92h
[21:47:37] step= 35700 | loss=4.1774 | acc30=0.039 | lr=1.00e-08 | 1.92h
[21:47:55] step= 35800 | loss=3.0845 | acc30=0.103 | lr=1.00e-08 | 1.93h
[21:48:13] step= 35900 | loss=2.2509 | acc30=0.263 | lr=1.00e-08 | 1.93h
[21:48:30] step= 36000 | loss=2.1601 | acc30=0.299 | lr=1.00e-08 | 1.94h
[21:48:30] Validando...
[21:48:47] VAL: loss=2.3549 acc30=0.200 acc60=0.272 acc120=0.337
[21:49:05] step= 36100 | loss=3.0824 | acc30=0.081 | lr=1.00e-08 | 1.94h
[21:49:23] step= 36200 | loss=2.8631 | acc30=0.107 | lr=1.00e-08 | 1.95h
[21:49:41] step= 36300 | loss=2.8420 | acc30=0.113 | lr=1.00e-08 | 1.95h
[21:49:58] step= 36400 | loss=5.5058 | acc30=0.002 | lr=1.00e-08 | 1.96h
[21:50:16] step= 36500 | loss=2.7960 | acc30=0.154 | lr=1.00e-08 | 1.96h
[21:50:34] step= 36600 | loss=2.2516 | acc30=0.245 | lr=1.00e-08 | 1.97h
[21:50:52] step= 36700 | loss=2.1723 | acc30=0.282 | lr=1.00e-08 | 1.97h
[21:51:10] step= 36800 | loss=2.8862 | acc30=0.115 | lr=1.00e-08 | 1.98h
[21:51:28] step= 36900 | loss=2.3265 | acc30=0.255 | lr=1.00e-08 | 1.98h
[21:51:46] step= 37000 | loss=2.2760 | acc30=0.251 | lr=1.00e-08 | 1.99h
[21:51:46] Validando...
[21:52:02] VAL: loss=2.3549 acc30=0.200 acc60=0.271 acc120=0.337
[21:52:20] step= 37100 | loss=2.5595 | acc30=0.128 | lr=1.00e-08 | 2.00h
[21:52:24] --- CHECKPOINT (hora 2.0) ---
[21:52:24] Checkpoint salvo: checkpoint_step00037125.pt (59.0 MB)
[21:52:31] ☁ Upload HuggingFace OK: step=37125
[21:52:38] step= 37200 | loss=2.4483 | acc30=0.123 | lr=1.00e-08 | 2.00h
[21:52:56] step= 37300 | loss=2.1902 | acc30=0.285 | lr=1.00e-08 | 2.01h
[21:53:14] step= 37400 | loss=2.1812 | acc30=0.316 | lr=1.00e-08 | 2.01h
[21:53:32] step= 37500 | loss=2.2150 | acc30=0.259 | lr=1.00e-08 | 2.02h
[21:53:49] step= 37600 | loss=2.1194 | acc30=0.315 | lr=1.00e-08 | 2.02h
[21:54:07] step= 37700 | loss=2.5962 | acc30=0.154 | lr=1.00e-08 | 2.03h
[21:54:25] step= 37800 | loss=2.6096 | acc30=0.125 | lr=1.00e-08 | 2.03h
[21:54:42] step= 37900 | loss=2.3120 | acc30=0.208 | lr=1.00e-08 | 2.04h
[21:55:00] step= 38000 | loss=2.1440 | acc30=0.309 | lr=1.00e-08 | 2.04h
[21:55:00] Validando...
[21:55:16] VAL: loss=2.3548 acc30=0.200 acc60=0.272 acc120=0.337
[21:55:34] step= 38100 | loss=3.0425 | acc30=0.139 | lr=1.00e-08 | 2.05h
[21:55:52] step= 38200 | loss=3.6513 | acc30=0.037 | lr=1.00e-08 | 2.06h
[21:56:09] step= 38300 | loss=2.3805 | acc30=0.191 | lr=1.00e-08 | 2.06h
[21:56:27] step= 38400 | loss=2.3209 | acc30=0.204 | lr=1.00e-08 | 2.07h
[21:56:45] step= 38500 | loss=2.1988 | acc30=0.291 | lr=1.00e-08 | 2.07h
[21:57:03] step= 38600 | loss=2.1534 | acc30=0.328 | lr=1.00e-08 | 2.08h
[21:57:20] step= 38700 | loss=2.2953 | acc30=0.228 | lr=1.00e-08 | 2.08h
[21:57:38] step= 38800 | loss=2.9650 | acc30=0.087 | lr=1.00e-08 | 2.09h
[21:57:56] step= 38900 | loss=2.7319 | acc30=0.150 | lr=1.00e-08 | 2.09h
[21:58:14] step= 39000 | loss=2.2150 | acc30=0.279 | lr=1.00e-08 | 2.10h
[21:58:14] Validando...
[21:58:30] VAL: loss=2.3548 acc30=0.200 acc60=0.272 acc120=0.337
[21:58:47] step= 39100 | loss=2.1509 | acc30=0.296 | lr=1.00e-08 | 2.11h
[21:59:06] step= 39200 | loss=3.6094 | acc30=0.079 | lr=1.00e-08 | 2.11h
[21:59:24] step= 39300 | loss=2.9615 | acc30=0.110 | lr=1.00e-08 | 2.12h
[21:59:41] step= 39400 | loss=2.1852 | acc30=0.277 | lr=1.00e-08 | 2.12h
[21:59:59] step= 39500 | loss=2.2194 | acc30=0.260 | lr=1.00e-08 | 2.13h
[22:00:17] step= 39600 | loss=2.3408 | acc30=0.192 | lr=1.00e-08 | 2.13h
[22:00:35] step= 39700 | loss=2.7859 | acc30=0.126 | lr=1.00e-08 | 2.14h
[22:00:53] step= 39800 | loss=2.9621 | acc30=0.083 | lr=1.00e-08 | 2.14h
[22:01:10] step= 39900 | loss=2.7538 | acc30=0.100 | lr=1.00e-08 | 2.15h
[22:01:28] step= 40000 | loss=2.4005 | acc30=0.179 | lr=1.00e-08 | 2.15h
[22:01:28] Validando...
[22:01:44] VAL: loss=2.3547 acc30=0.200 acc60=0.272 acc120=0.337
[22:02:02] step= 40100 | loss=2.2993 | acc30=0.214 | lr=1.00e-08 | 2.16h
[22:02:20] step= 40200 | loss=2.2439 | acc30=0.244 | lr=1.00e-08 | 2.17h
[22:02:37] step= 40300 | loss=2.2034 | acc30=0.262 | lr=1.00e-08 | 2.17h
[22:02:55] step= 40400 | loss=2.2572 | acc30=0.203 | lr=1.00e-08 | 2.18h
[22:03:13] step= 40500 | loss=2.5325 | acc30=0.159 | lr=1.00e-08 | 2.18h
[22:03:31] step= 40600 | loss=2.8841 | acc30=0.093 | lr=1.00e-08 | 2.19h
[22:03:48] step= 40700 | loss=2.8165 | acc30=0.130 | lr=1.00e-08 | 2.19h
[22:04:06] step= 40800 | loss=3.8444 | acc30=0.044 | lr=1.00e-08 | 2.20h
[22:04:24] step= 40900 | loss=2.9210 | acc30=0.166 | lr=1.00e-08 | 2.20h
[22:04:42] step= 41000 | loss=2.0805 | acc30=0.368 | lr=1.00e-08 | 2.21h
[22:04:42] Validando...
[22:04:58] VAL: loss=2.3547 acc30=0.200 acc60=0.272 acc120=0.337
[22:05:15] step= 41100 | loss=2.1406 | acc30=0.303 | lr=1.00e-08 | 2.21h
[22:05:33] step= 41200 | loss=2.2848 | acc30=0.221 | lr=1.00e-08 | 2.22h
[22:05:51] step= 41300 | loss=2.8379 | acc30=0.151 | lr=1.00e-08 | 2.22h
[22:06:09] step= 41400 | loss=2.8038 | acc30=0.101 | lr=1.00e-08 | 2.23h
[22:06:27] step= 41500 | loss=3.8670 | acc30=0.020 | lr=1.00e-08 | 2.23h
[22:06:45] step= 41600 | loss=2.8089 | acc30=0.175 | lr=1.00e-08 | 2.24h
[22:07:03] step= 41700 | loss=2.2890 | acc30=0.234 | lr=1.00e-08 | 2.24h
[22:07:20] step= 41800 | loss=2.3442 | acc30=0.188 | lr=1.00e-08 | 2.25h
[22:07:38] step= 41900 | loss=2.3151 | acc30=0.206 | lr=1.00e-08 | 2.25h
[22:07:56] step= 42000 | loss=2.1629 | acc30=0.279 | lr=1.00e-08 | 2.26h
[22:07:56] Validando...
[22:08:12] VAL: loss=2.3546 acc30=0.200 acc60=0.272 acc120=0.337
[22:08:30] step= 42100 | loss=2.4272 | acc30=0.228 | lr=1.00e-08 | 2.27h
[22:08:47] step= 42200 | loss=2.8240 | acc30=0.107 | lr=1.00e-08 | 2.27h
[22:09:05] step= 42300 | loss=2.3849 | acc30=0.190 | lr=1.00e-08 | 2.28h
[22:09:23] step= 42400 | loss=2.1819 | acc30=0.282 | lr=1.00e-08 | 2.28h
[22:09:41] step= 42500 | loss=2.1987 | acc30=0.284 | lr=1.00e-08 | 2.29h
[22:09:58] step= 42600 | loss=2.7702 | acc30=0.096 | lr=1.00e-08 | 2.29h
[22:10:16] step= 42700 | loss=3.6808 | acc30=0.033 | lr=1.00e-08 | 2.30h
[22:10:34] step= 42800 | loss=2.9999 | acc30=0.118 | lr=1.00e-08 | 2.30h
[22:10:52] step= 42900 | loss=2.3242 | acc30=0.235 | lr=1.00e-08 | 2.31h
[22:11:10] step= 43000 | loss=2.6670 | acc30=0.114 | lr=1.00e-08 | 2.31h
[22:11:10] Validando...
[22:11:26] VAL: loss=2.3546 acc30=0.200 acc60=0.272 acc120=0.337
[22:11:44] step= 43100 | loss=2.3373 | acc30=0.201 | lr=1.00e-08 | 2.32h
[22:12:02] step= 43200 | loss=2.2648 | acc30=0.238 | lr=1.00e-08 | 2.33h
[22:12:20] step= 43300 | loss=2.4278 | acc30=0.186 | lr=1.00e-08 | 2.33h
[22:12:37] step= 43400 | loss=2.7307 | acc30=0.119 | lr=1.00e-08 | 2.34h
[22:12:55] step= 43500 | loss=2.2343 | acc30=0.248 | lr=1.00e-08 | 2.34h
[22:13:13] step= 43600 | loss=2.1800 | acc30=0.274 | lr=1.00e-08 | 2.35h
[22:13:31] step= 43700 | loss=2.1821 | acc30=0.265 | lr=1.00e-08 | 2.35h
[22:13:48] step= 43800 | loss=2.1672 | acc30=0.288 | lr=1.00e-08 | 2.36h
[22:14:06] step= 43900 | loss=2.1912 | acc30=0.260 | lr=1.00e-08 | 2.36h
[22:14:24] step= 44000 | loss=2.6218 | acc30=0.161 | lr=1.00e-08 | 2.37h
[22:14:24] Validando...
[22:14:40] VAL: loss=2.3545 acc30=0.200 acc60=0.272 acc120=0.337
[22:14:58] step= 44100 | loss=2.6906 | acc30=0.111 | lr=1.00e-08 | 2.38h
[22:15:16] step= 44200 | loss=2.4834 | acc30=0.141 | lr=1.00e-08 | 2.38h
[22:15:34] step= 44300 | loss=2.1908 | acc30=0.259 | lr=1.00e-08 | 2.39h
[22:15:52] step= 44400 | loss=2.3245 | acc30=0.202 | lr=1.00e-08 | 2.39h
[22:16:10] step= 44500 | loss=2.4851 | acc30=0.149 | lr=1.00e-08 | 2.40h
[22:16:27] step= 44600 | loss=2.5626 | acc30=0.136 | lr=1.00e-08 | 2.40h
[22:16:45] step= 44700 | loss=4.5400 | acc30=0.025 | lr=1.00e-08 | 2.41h
[22:17:03] step= 44800 | loss=3.8161 | acc30=0.096 | lr=1.00e-08 | 2.41h
[22:17:21] step= 44900 | loss=2.1596 | acc30=0.305 | lr=1.00e-08 | 2.42h
[22:17:38] step= 45000 | loss=2.1860 | acc30=0.282 | lr=1.00e-08 | 2.42h
[22:17:38] Validando...
[22:17:54] VAL: loss=2.3545 acc30=0.200 acc60=0.272 acc120=0.337
[22:18:13] step= 45100 | loss=3.0106 | acc30=0.108 | lr=1.00e-08 | 2.43h
[22:18:31] step= 45200 | loss=2.5305 | acc30=0.161 | lr=1.00e-08 | 2.44h
[22:18:49] step= 45300 | loss=2.4597 | acc30=0.174 | lr=1.00e-08 | 2.44h
[22:19:06] step= 45400 | loss=2.1496 | acc30=0.280 | lr=1.00e-08 | 2.44h
[22:19:24] step= 45500 | loss=2.2757 | acc30=0.222 | lr=1.00e-08 | 2.45h
[22:19:41] step= 45600 | loss=2.2510 | acc30=0.254 | lr=1.00e-08 | 2.45h
[22:19:59] step= 45700 | loss=2.6526 | acc30=0.127 | lr=1.00e-08 | 2.46h
[22:20:17] step= 45800 | loss=3.0973 | acc30=0.074 | lr=1.00e-08 | 2.46h
[22:20:34] step= 45900 | loss=2.3744 | acc30=0.216 | lr=1.00e-08 | 2.47h
[22:20:52] step= 46000 | loss=2.3055 | acc30=0.224 | lr=1.00e-08 | 2.47h
[22:20:52] Validando...
[22:21:09] VAL: loss=2.3544 acc30=0.200 acc60=0.272 acc120=0.337
[22:21:27] step= 46100 | loss=2.5243 | acc30=0.133 | lr=1.00e-08 | 2.48h
[22:21:44] step= 46200 | loss=2.7369 | acc30=0.091 | lr=1.00e-08 | 2.49h
[22:22:02] step= 46300 | loss=2.2550 | acc30=0.302 | lr=1.00e-08 | 2.49h
[22:22:20] step= 46400 | loss=2.1036 | acc30=0.353 | lr=1.00e-08 | 2.50h
[22:22:38] step= 46500 | loss=2.0897 | acc30=0.353 | lr=1.00e-08 | 2.50h
[22:22:55] step= 46600 | loss=2.0725 | acc30=0.404 | lr=1.00e-08 | 2.51h
[22:23:13] step= 46700 | loss=2.2472 | acc30=0.241 | lr=1.00e-08 | 2.51h
[22:23:31] step= 46800 | loss=2.2262 | acc30=0.245 | lr=1.00e-08 | 2.52h
[22:23:49] step= 46900 | loss=2.1710 | acc30=0.325 | lr=1.00e-08 | 2.52h
[22:24:06] step= 47000 | loss=3.1587 | acc30=0.060 | lr=1.00e-08 | 2.53h
[22:24:06] Validando...
[22:24:22] VAL: loss=2.3544 acc30=0.200 acc60=0.272 acc120=0.337
[22:24:40] step= 47100 | loss=2.3117 | acc30=0.244 | lr=1.00e-08 | 2.54h
[22:24:59] step= 47200 | loss=2.2372 | acc30=0.262 | lr=1.00e-08 | 2.54h
[22:25:16] step= 47300 | loss=2.3251 | acc30=0.208 | lr=1.00e-08 | 2.55h
[22:25:34] step= 47400 | loss=2.3847 | acc30=0.178 | lr=1.00e-08 | 2.55h
[22:25:52] step= 47500 | loss=2.3791 | acc30=0.182 | lr=1.00e-08 | 2.56h
[22:26:10] step= 47600 | loss=2.3162 | acc30=0.252 | lr=1.00e-08 | 2.56h
[22:26:28] step= 47700 | loss=2.1495 | acc30=0.316 | lr=1.00e-08 | 2.57h
[22:26:45] step= 47800 | loss=2.3029 | acc30=0.231 | lr=1.00e-08 | 2.57h
[22:27:03] step= 47900 | loss=2.4236 | acc30=0.179 | lr=1.00e-08 | 2.58h
[22:27:21] step= 48000 | loss=2.3097 | acc30=0.209 | lr=1.00e-08 | 2.58h
[22:27:21] Validando...
[22:27:37] VAL: loss=2.3543 acc30=0.200 acc60=0.271 acc120=0.337
[22:27:55] step= 48100 | loss=2.1989 | acc30=0.275 | lr=1.00e-08 | 2.59h
[22:28:13] step= 48200 | loss=2.2950 | acc30=0.213 | lr=1.00e-08 | 2.60h
[22:28:31] step= 48300 | loss=2.8182 | acc30=0.107 | lr=1.00e-08 | 2.60h
[22:28:49] step= 48400 | loss=2.5638 | acc30=0.160 | lr=1.00e-08 | 2.61h
[22:29:06] step= 48500 | loss=2.2309 | acc30=0.252 | lr=1.00e-08 | 2.61h
[22:29:24] step= 48600 | loss=2.4715 | acc30=0.158 | lr=1.00e-08 | 2.62h
[22:29:42] step= 48700 | loss=2.3785 | acc30=0.185 | lr=1.00e-08 | 2.62h
[22:30:00] step= 48800 | loss=2.2623 | acc30=0.226 | lr=1.00e-08 | 2.63h
[22:30:17] step= 48900 | loss=2.1020 | acc30=0.332 | lr=1.00e-08 | 2.63h
[22:30:35] step= 49000 | loss=2.4658 | acc30=0.150 | lr=1.00e-08 | 2.64h
[22:30:35] Validando...
[22:30:52] VAL: loss=2.3543 acc30=0.200 acc60=0.272 acc120=0.337
[22:31:10] step= 49100 | loss=2.3197 | acc30=0.218 | lr=1.00e-08 | 2.65h
[22:31:28] step= 49200 | loss=2.2015 | acc30=0.267 | lr=1.00e-08 | 2.65h
[22:31:45] step= 49300 | loss=2.2209 | acc30=0.232 | lr=1.00e-08 | 2.66h
[22:32:03] step= 49400 | loss=2.2736 | acc30=0.192 | lr=1.00e-08 | 2.66h
[22:32:21] step= 49500 | loss=2.2987 | acc30=0.205 | lr=1.00e-08 | 2.67h
[22:32:40] step= 49600 | loss=2.4144 | acc30=0.171 | lr=1.00e-08 | 2.67h
[22:32:57] step= 49700 | loss=2.2434 | acc30=0.236 | lr=1.00e-08 | 2.68h
[22:33:15] step= 49800 | loss=2.3625 | acc30=0.185 | lr=1.00e-08 | 2.68h
[22:33:33] step= 49900 | loss=2.4361 | acc30=0.170 | lr=1.00e-08 | 2.69h
[22:33:51] step= 50000 | loss=2.4214 | acc30=0.166 | lr=1.00e-08 | 2.69h
[22:33:51] Validando...
[22:34:07] VAL: loss=2.3542 acc30=0.200 acc60=0.271 acc120=0.337
[22:34:24] step= 50100 | loss=2.7815 | acc30=0.093 | lr=1.00e-08 | 2.70h
[22:34:42] step= 50200 | loss=2.2762 | acc30=0.228 | lr=1.00e-08 | 2.71h
[22:35:00] step= 50300 | loss=2.2170 | acc30=0.250 | lr=1.00e-08 | 2.71h
[22:35:18] step= 50400 | loss=2.1643 | acc30=0.292 | lr=1.00e-08 | 2.72h
[22:35:36] step= 50500 | loss=2.0821 | acc30=0.380 | lr=1.00e-08 | 2.72h
[22:35:54] step= 50600 | loss=2.1190 | acc30=0.327 | lr=1.00e-08 | 2.72h
[22:36:12] step= 50700 | loss=2.1617 | acc30=0.323 | lr=1.00e-08 | 2.73h
[22:36:30] step= 50800 | loss=2.2845 | acc30=0.293 | lr=1.00e-08 | 2.73h
[22:36:47] step= 50900 | loss=3.4290 | acc30=0.047 | lr=1.00e-08 | 2.74h
[22:37:05] step= 51000 | loss=2.5700 | acc30=0.198 | lr=1.00e-08 | 2.74h
[22:37:05] Validando...
[22:37:21] VAL: loss=2.3542 acc30=0.200 acc60=0.271 acc120=0.337
[22:37:39] step= 51100 | loss=2.1913 | acc30=0.276 | lr=1.00e-08 | 2.75h
[22:37:57] step= 51200 | loss=2.4712 | acc30=0.177 | lr=1.00e-08 | 2.76h
[22:38:14] step= 51300 | loss=4.6946 | acc30=0.009 | lr=1.00e-08 | 2.76h
[22:38:32] step= 51400 | loss=3.0514 | acc30=0.091 | lr=1.00e-08 | 2.77h
[22:38:50] step= 51500 | loss=2.1602 | acc30=0.311 | lr=1.00e-08 | 2.77h
[22:39:08] step= 51600 | loss=2.1014 | acc30=0.364 | lr=1.00e-08 | 2.78h
[22:39:26] step= 51700 | loss=3.0179 | acc30=0.083 | lr=1.00e-08 | 2.78h
[22:39:44] step= 51800 | loss=2.2771 | acc30=0.222 | lr=1.00e-08 | 2.79h
[22:40:02] step= 51900 | loss=2.4889 | acc30=0.167 | lr=1.00e-08 | 2.79h
[22:40:20] step= 52000 | loss=2.6624 | acc30=0.119 | lr=1.00e-08 | 2.80h
[22:40:20] Validando...
[22:40:36] VAL: loss=2.3542 acc30=0.200 acc60=0.272 acc120=0.337
[22:40:53] step= 52100 | loss=2.2309 | acc30=0.245 | lr=1.00e-08 | 2.81h
[22:41:11] step= 52200 | loss=2.1230 | acc30=0.321 | lr=1.00e-08 | 2.81h
[22:41:29] step= 52300 | loss=2.2463 | acc30=0.271 | lr=1.00e-08 | 2.82h
[22:41:47] step= 52400 | loss=2.8453 | acc30=0.092 | lr=1.00e-08 | 2.82h
[22:42:04] step= 52500 | loss=2.0791 | acc30=0.380 | lr=1.00e-08 | 2.83h
[22:42:23] step= 52600 | loss=2.1769 | acc30=0.288 | lr=1.00e-08 | 2.83h
[22:42:40] step= 52700 | loss=2.2262 | acc30=0.244 | lr=1.00e-08 | 2.84h
[22:42:58] step= 52800 | loss=2.2347 | acc30=0.240 | lr=1.00e-08 | 2.84h
[22:43:16] step= 52900 | loss=2.2348 | acc30=0.254 | lr=1.00e-08 | 2.85h
[22:43:34] step= 53000 | loss=2.5873 | acc30=0.164 | lr=1.00e-08 | 2.85h
[22:43:34] Validando...
[22:43:50] VAL: loss=2.3541 acc30=0.200 acc60=0.272 acc120=0.337
[22:44:08] step= 53100 | loss=2.4186 | acc30=0.179 | lr=1.00e-08 | 2.86h
[22:44:25] step= 53200 | loss=2.1302 | acc30=0.322 | lr=1.00e-08 | 2.87h
[22:44:43] step= 53300 | loss=2.3996 | acc30=0.160 | lr=1.00e-08 | 2.87h
[22:45:01] step= 53400 | loss=3.4951 | acc30=0.062 | lr=1.00e-08 | 2.88h
[22:45:19] step= 53500 | loss=3.1804 | acc30=0.102 | lr=1.00e-08 | 2.88h
[22:45:36] step= 53600 | loss=2.1377 | acc30=0.315 | lr=1.00e-08 | 2.89h
[22:45:54] step= 53700 | loss=2.4673 | acc30=0.173 | lr=1.00e-08 | 2.89h
[22:46:12] step= 53800 | loss=2.3813 | acc30=0.189 | lr=1.00e-08 | 2.90h
[22:46:30] step= 53900 | loss=2.3624 | acc30=0.200 | lr=1.00e-08 | 2.90h
[22:46:47] step= 54000 | loss=2.6613 | acc30=0.112 | lr=1.00e-08 | 2.91h
[22:46:47] Validando...
[22:47:03] VAL: loss=2.3541 acc30=0.200 acc60=0.272 acc120=0.337
[22:47:21] step= 54100 | loss=2.2447 | acc30=0.289 | lr=1.00e-08 | 2.92h
[22:47:40] step= 54200 | loss=3.7215 | acc30=0.166 | lr=1.00e-08 | 2.92h
[22:47:58] step= 54300 | loss=5.1542 | acc30=0.055 | lr=1.00e-08 | 2.93h
[22:48:15] step= 54400 | loss=2.1289 | acc30=0.315 | lr=1.00e-08 | 2.93h
[22:48:33] step= 54500 | loss=2.1930 | acc30=0.270 | lr=1.00e-08 | 2.94h
[22:48:51] step= 54600 | loss=2.1953 | acc30=0.263 | lr=1.00e-08 | 2.94h
[22:49:09] step= 54700 | loss=2.1750 | acc30=0.290 | lr=1.00e-08 | 2.95h
[22:49:26] step= 54800 | loss=2.1632 | acc30=0.281 | lr=1.00e-08 | 2.95h
[22:49:44] step= 54900 | loss=2.2679 | acc30=0.226 | lr=1.00e-08 | 2.96h
[22:50:02] step= 55000 | loss=2.2276 | acc30=0.270 | lr=1.00e-08 | 2.96h
[22:50:02] Validando...
[22:50:18] VAL: loss=2.3540 acc30=0.200 acc60=0.272 acc120=0.337
[22:50:36] step= 55100 | loss=2.2702 | acc30=0.225 | lr=1.00e-08 | 2.97h
[22:50:53] step= 55200 | loss=2.1556 | acc30=0.283 | lr=1.00e-08 | 2.97h
[22:51:11] step= 55300 | loss=2.6974 | acc30=0.134 | lr=1.00e-08 | 2.98h
[22:51:29] step= 55400 | loss=2.2555 | acc30=0.235 | lr=1.00e-08 | 2.98h
[22:51:47] step= 55500 | loss=2.1963 | acc30=0.250 | lr=1.00e-08 | 2.99h
[22:52:04] step= 55600 | loss=2.3856 | acc30=0.189 | lr=1.00e-08 | 2.99h
[22:52:22] step= 55700 | loss=3.2295 | acc30=0.085 | lr=1.00e-08 | 3.00h
[22:52:24] --- CHECKPOINT (hora 3.0) ---
[22:52:24] Checkpoint salvo: checkpoint_step00055712.pt (59.0 MB)
[22:52:30] ☁ Upload HuggingFace OK: step=55712
[22:52:41] step= 55800 | loss=2.9909 | acc30=0.156 | lr=1.00e-08 | 3.00h
[22:52:59] step= 55900 | loss=2.0742 | acc30=0.397 | lr=1.00e-08 | 3.01h
[22:53:16] step= 56000 | loss=4.4831 | acc30=0.021 | lr=1.00e-08 | 3.01h
[22:53:16] Validando...
[22:53:32] VAL: loss=2.3540 acc30=0.200 acc60=0.271 acc120=0.337
[22:53:50] step= 56100 | loss=2.3610 | acc30=0.193 | lr=1.00e-08 | 3.02h
[22:54:09] step= 56200 | loss=2.2089 | acc30=0.252 | lr=1.00e-08 | 3.03h
[22:54:26] step= 56300 | loss=2.1536 | acc30=0.295 | lr=1.00e-08 | 3.03h
[22:54:44] step= 56400 | loss=2.2717 | acc30=0.228 | lr=1.00e-08 | 3.04h
[22:55:02] step= 56500 | loss=2.2863 | acc30=0.217 | lr=1.00e-08 | 3.04h
[22:55:20] step= 56600 | loss=2.3772 | acc30=0.219 | lr=1.00e-08 | 3.05h
[22:55:37] step= 56700 | loss=2.7942 | acc30=0.112 | lr=1.00e-08 | 3.05h
[22:55:55] step= 56800 | loss=2.2847 | acc30=0.240 | lr=1.00e-08 | 3.06h
[22:56:13] step= 56900 | loss=2.2913 | acc30=0.220 | lr=1.00e-08 | 3.06h
[22:56:32] step= 57000 | loss=2.3060 | acc30=0.236 | lr=1.00e-08 | 3.07h
[22:56:32] Validando...
[22:56:48] VAL: loss=2.3539 acc30=0.200 acc60=0.271 acc120=0.337
[22:57:05] step= 57100 | loss=2.1354 | acc30=0.314 | lr=1.00e-08 | 3.08h
[22:57:23] step= 57200 | loss=2.1181 | acc30=0.340 | lr=1.00e-08 | 3.08h
[22:57:41] step= 57300 | loss=2.4983 | acc30=0.205 | lr=1.00e-08 | 3.09h
[22:57:59] step= 57400 | loss=2.5378 | acc30=0.175 | lr=1.00e-08 | 3.09h
[22:58:17] step= 57500 | loss=2.1528 | acc30=0.309 | lr=1.00e-08 | 3.10h
[22:58:35] step= 57600 | loss=2.2481 | acc30=0.233 | lr=1.00e-08 | 3.10h
[22:58:53] step= 57700 | loss=2.0927 | acc30=0.351 | lr=1.00e-08 | 3.11h
[22:59:11] step= 57800 | loss=2.1765 | acc30=0.330 | lr=1.00e-08 | 3.11h
[22:59:28] step= 57900 | loss=2.4018 | acc30=0.187 | lr=1.00e-08 | 3.12h
[22:59:46] step= 58000 | loss=2.1278 | acc30=0.337 | lr=1.00e-08 | 3.12h
[22:59:46] Validando...
[23:00:02] VAL: loss=2.3539 acc30=0.200 acc60=0.272 acc120=0.337
[23:00:20] step= 58100 | loss=2.3080 | acc30=0.236 | lr=1.00e-08 | 3.13h
[23:00:39] step= 58200 | loss=2.1964 | acc30=0.266 | lr=1.00e-08 | 3.14h
[23:00:56] step= 58300 | loss=2.6340 | acc30=0.120 | lr=1.00e-08 | 3.14h
[23:01:14] step= 58400 | loss=2.2567 | acc30=0.289 | lr=1.00e-08 | 3.15h
[23:01:32] step= 58500 | loss=2.1296 | acc30=0.329 | lr=1.00e-08 | 3.15h
[23:01:50] step= 58600 | loss=2.2153 | acc30=0.287 | lr=1.00e-08 | 3.16h
[23:02:07] step= 58700 | loss=2.7896 | acc30=0.087 | lr=1.00e-08 | 3.16h
[23:02:25] step= 58800 | loss=3.7694 | acc30=0.030 | lr=1.00e-08 | 3.17h
[23:02:43] step= 58900 | loss=2.9895 | acc30=0.172 | lr=1.00e-08 | 3.17h
[23:03:01] step= 59000 | loss=2.1703 | acc30=0.292 | lr=1.00e-08 | 3.18h
[23:03:01] Validando...
[23:03:17] VAL: loss=2.3538 acc30=0.200 acc60=0.272 acc120=0.337
[23:03:34] step= 59100 | loss=2.2143 | acc30=0.257 | lr=1.00e-08 | 3.19h
[23:03:52] step= 59200 | loss=2.1340 | acc30=0.316 | lr=1.00e-08 | 3.19h
[23:04:10] step= 59300 | loss=2.6779 | acc30=0.212 | lr=1.00e-08 | 3.20h
[23:04:28] step= 59400 | loss=2.7009 | acc30=0.130 | lr=1.00e-08 | 3.20h
[23:04:45] step= 59500 | loss=2.1256 | acc30=0.322 | lr=1.00e-08 | 3.21h
[23:05:03] step= 59600 | loss=2.2170 | acc30=0.258 | lr=1.00e-08 | 3.21h
[23:05:21] step= 59700 | loss=2.2213 | acc30=0.241 | lr=1.00e-08 | 3.22h
[23:05:39] step= 59800 | loss=2.2089 | acc30=0.236 | lr=1.00e-08 | 3.22h
[23:05:56] step= 59900 | loss=2.2164 | acc30=0.245 | lr=1.00e-08 | 3.23h
[23:06:14] step= 60000 | loss=2.3375 | acc30=0.191 | lr=1.00e-08 | 3.23h
[23:06:14] Validando...
[23:06:30] VAL: loss=2.3538 acc30=0.200 acc60=0.272 acc120=0.337
[23:06:48] step= 60100 | loss=2.2735 | acc30=0.224 | lr=1.00e-08 | 3.24h
[23:07:06] step= 60200 | loss=2.3135 | acc30=0.207 | lr=1.00e-08 | 3.25h
[23:07:23] step= 60300 | loss=2.2916 | acc30=0.198 | lr=1.00e-08 | 3.25h
[23:07:41] step= 60400 | loss=2.4599 | acc30=0.157 | lr=1.00e-08 | 3.25h
[23:07:59] step= 60500 | loss=2.1560 | acc30=0.288 | lr=1.00e-08 | 3.26h
[23:08:17] step= 60600 | loss=2.2581 | acc30=0.231 | lr=1.00e-08 | 3.26h
[23:08:34] step= 60700 | loss=3.3693 | acc30=0.065 | lr=1.00e-08 | 3.27h
[23:08:52] step= 60800 | loss=3.0105 | acc30=0.100 | lr=1.00e-08 | 3.27h
[23:09:10] step= 60900 | loss=3.0241 | acc30=0.099 | lr=1.00e-08 | 3.28h
[23:09:28] step= 61000 | loss=2.2564 | acc30=0.238 | lr=1.00e-08 | 3.28h
[23:09:28] Validando...
[23:09:45] VAL: loss=2.3538 acc30=0.200 acc60=0.272 acc120=0.337
[23:10:02] step= 61100 | loss=2.3111 | acc30=0.218 | lr=1.00e-08 | 3.29h
[23:10:21] step= 61200 | loss=2.4487 | acc30=0.179 | lr=1.00e-08 | 3.30h
[23:10:38] step= 61300 | loss=2.3120 | acc30=0.238 | lr=1.00e-08 | 3.30h
[23:10:56] step= 61400 | loss=2.4051 | acc30=0.170 | lr=1.00e-08 | 3.31h
[23:11:13] step= 61500 | loss=3.0412 | acc30=0.076 | lr=1.00e-08 | 3.31h
[23:11:31] step= 61600 | loss=2.7332 | acc30=0.240 | lr=1.00e-08 | 3.32h
[23:11:49] step= 61700 | loss=2.1831 | acc30=0.332 | lr=1.00e-08 | 3.32h
[23:12:06] step= 61800 | loss=3.4265 | acc30=0.042 | lr=1.00e-08 | 3.33h
[23:12:24] step= 61900 | loss=2.3512 | acc30=0.215 | lr=1.00e-08 | 3.33h
[23:12:42] step= 62000 | loss=2.1792 | acc30=0.278 | lr=1.00e-08 | 3.34h
[23:12:42] Validando...
[23:12:58] VAL: loss=2.3537 acc30=0.200 acc60=0.272 acc120=0.337
[23:13:16] step= 62100 | loss=2.1671 | acc30=0.287 | lr=1.00e-08 | 3.35h
[23:13:34] step= 62200 | loss=2.3053 | acc30=0.227 | lr=1.00e-08 | 3.35h
[23:13:51] step= 62300 | loss=2.3611 | acc30=0.173 | lr=1.00e-08 | 3.36h
[23:14:09] step= 62400 | loss=2.3613 | acc30=0.177 | lr=1.00e-08 | 3.36h
[23:14:27] step= 62500 | loss=2.4613 | acc30=0.174 | lr=1.00e-08 | 3.37h
[23:14:44] step= 62600 | loss=4.4345 | acc30=0.010 | lr=1.00e-08 | 3.37h
[23:15:02] step= 62700 | loss=4.4721 | acc30=0.015 | lr=1.00e-08 | 3.38h
[23:15:20] step= 62800 | loss=3.6790 | acc30=0.094 | lr=1.00e-08 | 3.38h
[23:15:38] step= 62900 | loss=2.3705 | acc30=0.193 | lr=1.00e-08 | 3.39h
[23:15:56] step= 63000 | loss=2.2634 | acc30=0.232 | lr=1.00e-08 | 3.39h
[23:15:56] Validando...
[23:16:12] VAL: loss=2.3537 acc30=0.200 acc60=0.272 acc120=0.337
[23:16:30] step= 63100 | loss=3.0016 | acc30=0.080 | lr=1.00e-08 | 3.40h
[23:16:48] step= 63200 | loss=2.8213 | acc30=0.117 | lr=1.00e-08 | 3.41h
[23:17:06] step= 63300 | loss=2.8651 | acc30=0.109 | lr=1.00e-08 | 3.41h
[23:17:23] step= 63400 | loss=3.2287 | acc30=0.056 | lr=1.00e-08 | 3.42h
[23:17:41] step= 63500 | loss=3.1651 | acc30=0.067 | lr=1.00e-08 | 3.42h
[23:17:59] step= 63600 | loss=2.7781 | acc30=0.135 | lr=1.00e-08 | 3.43h
[23:18:17] step= 63700 | loss=2.1250 | acc30=0.324 | lr=1.00e-08 | 3.43h
[23:18:34] step= 63800 | loss=2.0288 | acc30=0.431 | lr=1.00e-08 | 3.44h
[23:18:52] step= 63900 | loss=2.1862 | acc30=0.270 | lr=1.00e-08 | 3.44h
[23:19:10] step= 64000 | loss=2.2578 | acc30=0.231 | lr=1.00e-08 | 3.45h
[23:19:10] Validando...
[23:19:26] VAL: loss=2.3537 acc30=0.200 acc60=0.272 acc120=0.337
[23:19:44] step= 64100 | loss=2.3196 | acc30=0.198 | lr=1.00e-08 | 3.46h
[23:20:02] step= 64200 | loss=4.4721 | acc30=0.013 | lr=1.00e-08 | 3.46h
[23:20:20] step= 64300 | loss=2.8901 | acc30=0.198 | lr=1.00e-08 | 3.47h
[23:20:38] step= 64400 | loss=2.4126 | acc30=0.220 | lr=1.00e-08 | 3.47h
[23:20:55] step= 64500 | loss=2.6992 | acc30=0.112 | lr=1.00e-08 | 3.48h
[23:21:13] step= 64600 | loss=2.3838 | acc30=0.185 | lr=1.00e-08 | 3.48h
[23:21:31] step= 64700 | loss=2.2339 | acc30=0.240 | lr=1.00e-08 | 3.49h
[23:21:49] step= 64800 | loss=2.2322 | acc30=0.269 | lr=1.00e-08 | 3.49h
[23:22:06] step= 64900 | loss=3.3390 | acc30=0.049 | lr=1.00e-08 | 3.50h
[23:22:24] step= 65000 | loss=2.1824 | acc30=0.264 | lr=1.00e-08 | 3.50h
[23:22:24] Validando...
[23:22:40] VAL: loss=2.3537 acc30=0.200 acc60=0.272 acc120=0.337
[23:22:58] step= 65100 | loss=3.3295 | acc30=0.125 | lr=1.00e-08 | 3.51h
[23:23:16] step= 65200 | loss=3.2426 | acc30=0.109 | lr=1.00e-08 | 3.51h
[23:23:33] step= 65300 | loss=2.1343 | acc30=0.327 | lr=1.00e-08 | 3.52h
[23:23:51] step= 65400 | loss=2.1204 | acc30=0.321 | lr=1.00e-08 | 3.52h
[23:24:09] step= 65500 | loss=2.1203 | acc30=0.327 | lr=1.00e-08 | 3.53h
[23:24:27] step= 65600 | loss=2.1497 | acc30=0.328 | lr=1.00e-08 | 3.53h
[23:24:44] step= 65700 | loss=2.4327 | acc30=0.167 | lr=1.00e-08 | 3.54h
[23:25:02] step= 65800 | loss=2.3122 | acc30=0.192 | lr=1.00e-08 | 3.54h
[23:25:20] step= 65900 | loss=2.3411 | acc30=0.199 | lr=1.00e-08 | 3.55h
[23:25:38] step= 66000 | loss=2.2726 | acc30=0.223 | lr=1.00e-08 | 3.55h
[23:25:38] Validando...
[23:25:54] VAL: loss=2.3537 acc30=0.200 acc60=0.271 acc120=0.337
[23:26:11] step= 66100 | loss=2.2644 | acc30=0.233 | lr=1.00e-08 | 3.56h
[23:26:29] step= 66200 | loss=2.2908 | acc30=0.229 | lr=1.00e-08 | 3.57h
[23:26:47] step= 66300 | loss=2.2408 | acc30=0.236 | lr=1.00e-08 | 3.57h
[23:27:05] step= 66400 | loss=2.5139 | acc30=0.150 | lr=1.00e-08 | 3.58h
[23:27:22] step= 66500 | loss=3.2218 | acc30=0.058 | lr=1.00e-08 | 3.58h
[23:27:40] step= 66600 | loss=2.7375 | acc30=0.116 | lr=1.00e-08 | 3.59h
[23:27:58] step= 66700 | loss=2.4877 | acc30=0.176 | lr=1.00e-08 | 3.59h
[23:28:16] step= 66800 | loss=2.2405 | acc30=0.242 | lr=1.00e-08 | 3.60h
[23:28:34] step= 66900 | loss=2.2588 | acc30=0.224 | lr=1.00e-08 | 3.60h
[23:28:52] step= 67000 | loss=3.2053 | acc30=0.065 | lr=1.00e-08 | 3.61h
[23:28:52] Validando...
[23:29:08] VAL: loss=2.3536 acc30=0.200 acc60=0.271 acc120=0.337
[23:29:26] step= 67100 | loss=2.3031 | acc30=0.225 | lr=1.00e-08 | 3.62h
[23:29:44] step= 67200 | loss=2.4148 | acc30=0.193 | lr=1.00e-08 | 3.62h
[23:30:01] step= 67300 | loss=3.9936 | acc30=0.025 | lr=1.00e-08 | 3.63h
[23:30:19] step= 67400 | loss=2.4836 | acc30=0.275 | lr=1.00e-08 | 3.63h
[23:30:37] step= 67500 | loss=2.1016 | acc30=0.357 | lr=1.00e-08 | 3.64h
[23:30:55] step= 67600 | loss=2.0761 | acc30=0.367 | lr=1.00e-08 | 3.64h
[23:31:12] step= 67700 | loss=2.6893 | acc30=0.105 | lr=1.00e-08 | 3.65h
[23:31:30] step= 67800 | loss=2.6904 | acc30=0.102 | lr=1.00e-08 | 3.65h
[23:31:48] step= 67900 | loss=2.5381 | acc30=0.190 | lr=1.00e-08 | 3.66h
[23:32:06] step= 68000 | loss=2.1795 | acc30=0.286 | lr=1.00e-08 | 3.66h
[23:32:06] Validando...
[23:32:22] VAL: loss=2.3536 acc30=0.200 acc60=0.271 acc120=0.337
[23:32:39] step= 68100 | loss=2.1938 | acc30=0.250 | lr=1.00e-08 | 3.67h
[23:32:57] step= 68200 | loss=2.1620 | acc30=0.293 | lr=1.00e-08 | 3.68h
[23:33:15] step= 68300 | loss=2.2143 | acc30=0.268 | lr=1.00e-08 | 3.68h
[23:33:32] step= 68400 | loss=2.3573 | acc30=0.205 | lr=1.00e-08 | 3.69h
[23:33:50] step= 68500 | loss=2.3138 | acc30=0.225 | lr=1.00e-08 | 3.69h
[23:34:08] step= 68600 | loss=3.0000 | acc30=0.101 | lr=1.00e-08 | 3.70h
[23:34:26] step= 68700 | loss=2.5193 | acc30=0.174 | lr=1.00e-08 | 3.70h
[23:34:44] step= 68800 | loss=2.2703 | acc30=0.223 | lr=1.00e-08 | 3.71h
[23:35:01] step= 68900 | loss=2.9359 | acc30=0.099 | lr=1.00e-08 | 3.71h
[23:35:19] step= 69000 | loss=2.3679 | acc30=0.244 | lr=1.00e-08 | 3.72h
[23:35:19] Validando...
[23:35:35] VAL: loss=2.3536 acc30=0.200 acc60=0.271 acc120=0.337
[23:35:53] step= 69100 | loss=2.1180 | acc30=0.343 | lr=1.00e-08 | 3.72h
[23:36:11] step= 69200 | loss=2.1652 | acc30=0.282 | lr=1.00e-08 | 3.73h
[23:36:29] step= 69300 | loss=2.0896 | acc30=0.374 | lr=1.00e-08 | 3.73h
[23:36:47] step= 69400 | loss=2.0949 | acc30=0.354 | lr=1.00e-08 | 3.74h
[23:37:04] step= 69500 | loss=2.0962 | acc30=0.342 | lr=1.00e-08 | 3.74h
[23:37:22] step= 69600 | loss=2.2294 | acc30=0.244 | lr=1.00e-08 | 3.75h
[23:37:40] step= 69700 | loss=2.1721 | acc30=0.299 | lr=1.00e-08 | 3.75h
[23:37:58] step= 69800 | loss=2.2693 | acc30=0.260 | lr=1.00e-08 | 3.76h
[23:38:16] step= 69900 | loss=2.2205 | acc30=0.264 | lr=1.00e-08 | 3.76h
[23:38:33] step= 70000 | loss=2.1133 | acc30=0.340 | lr=1.00e-08 | 3.77h
[23:38:33] Validando...
[23:38:49] VAL: loss=2.3536 acc30=0.200 acc60=0.271 acc120=0.337
[23:39:07] step= 70100 | loss=3.0523 | acc30=0.096 | lr=1.00e-08 | 3.78h
[23:39:25] step= 70200 | loss=2.4662 | acc30=0.158 | lr=1.00e-08 | 3.78h
[23:39:43] step= 70300 | loss=2.2519 | acc30=0.246 | lr=1.00e-08 | 3.79h
[23:40:00] step= 70400 | loss=2.2048 | acc30=0.266 | lr=1.00e-08 | 3.79h
[23:40:18] step= 70500 | loss=2.2689 | acc30=0.209 | lr=1.00e-08 | 3.80h
[23:40:36] step= 70600 | loss=2.4044 | acc30=0.179 | lr=1.00e-08 | 3.80h
[23:40:54] step= 70700 | loss=2.3477 | acc30=0.224 | lr=1.00e-08 | 3.81h
[23:41:11] step= 70800 | loss=2.3079 | acc30=0.210 | lr=1.00e-08 | 3.81h
[23:41:29] step= 70900 | loss=3.2465 | acc30=0.081 | lr=1.00e-08 | 3.82h
[23:41:47] step= 71000 | loss=2.6231 | acc30=0.124 | lr=1.00e-08 | 3.82h
[23:41:47] Validando...
[23:42:03] VAL: loss=2.3536 acc30=0.200 acc60=0.272 acc120=0.337
[23:42:21] step= 71100 | loss=2.2700 | acc30=0.219 | lr=1.00e-08 | 3.83h
[23:42:39] step= 71200 | loss=2.3385 | acc30=0.200 | lr=1.00e-08 | 3.84h
[23:42:57] step= 71300 | loss=2.1661 | acc30=0.308 | lr=1.00e-08 | 3.84h
[23:43:14] step= 71400 | loss=3.0915 | acc30=0.242 | lr=1.00e-08 | 3.85h
[23:43:32] step= 71500 | loss=4.0131 | acc30=0.060 | lr=1.00e-08 | 3.85h
[23:43:49] step= 71600 | loss=2.0627 | acc30=0.388 | lr=1.00e-08 | 3.86h
[23:44:08] step= 71700 | loss=2.3261 | acc30=0.224 | lr=1.00e-08 | 3.86h
[23:44:26] step= 71800 | loss=2.2657 | acc30=0.229 | lr=1.00e-08 | 3.87h
[23:44:44] step= 71900 | loss=2.1852 | acc30=0.283 | lr=1.00e-08 | 3.87h
[23:45:01] step= 72000 | loss=2.6983 | acc30=0.101 | lr=1.00e-08 | 3.88h
[23:45:01] Validando...
[23:45:17] VAL: loss=2.3535 acc30=0.200 acc60=0.271 acc120=0.337
[23:45:35] step= 72100 | loss=2.3227 | acc30=0.208 | lr=1.00e-08 | 3.89h
[23:45:53] step= 72200 | loss=2.2166 | acc30=0.267 | lr=1.00e-08 | 3.89h
[23:46:11] step= 72300 | loss=2.3187 | acc30=0.213 | lr=1.00e-08 | 3.90h
[23:46:28] step= 72400 | loss=2.0820 | acc30=0.368 | lr=1.00e-08 | 3.90h
[23:46:46] step= 72500 | loss=2.3422 | acc30=0.217 | lr=1.00e-08 | 3.91h
[23:47:04] step= 72600 | loss=2.2990 | acc30=0.209 | lr=1.00e-08 | 3.91h
[23:47:21] step= 72700 | loss=2.4013 | acc30=0.161 | lr=1.00e-08 | 3.92h
[23:47:39] step= 72800 | loss=2.2030 | acc30=0.266 | lr=1.00e-08 | 3.92h
[23:47:56] step= 72900 | loss=2.6753 | acc30=0.140 | lr=1.00e-08 | 3.93h
[23:48:14] step= 73000 | loss=2.8032 | acc30=0.106 | lr=1.00e-08 | 3.93h
[23:48:14] Validando...
[23:48:30] VAL: loss=2.3535 acc30=0.200 acc60=0.271 acc120=0.337
[23:48:48] step= 73100 | loss=2.3250 | acc30=0.204 | lr=1.00e-08 | 3.94h
[23:49:06] step= 73200 | loss=2.7735 | acc30=0.100 | lr=1.00e-08 | 3.94h
[23:49:24] step= 73300 | loss=2.7989 | acc30=0.133 | lr=1.00e-08 | 3.95h
[23:49:41] step= 73400 | loss=3.0874 | acc30=0.105 | lr=1.00e-08 | 3.95h
[23:49:59] step= 73500 | loss=3.7177 | acc30=0.051 | lr=1.00e-08 | 3.96h
[23:50:17] step= 73600 | loss=2.1981 | acc30=0.342 | lr=1.00e-08 | 3.96h
[23:50:34] step= 73700 | loss=2.2466 | acc30=0.264 | lr=1.00e-08 | 3.97h
[23:50:53] step= 73800 | loss=2.2837 | acc30=0.216 | lr=1.00e-08 | 3.97h
[23:51:10] step= 73900 | loss=2.3517 | acc30=0.193 | lr=1.00e-08 | 3.98h
[23:51:28] step= 74000 | loss=2.0779 | acc30=0.374 | lr=1.00e-08 | 3.98h
[23:51:28] Validando...
[23:51:44] VAL: loss=2.3535 acc30=0.200 acc60=0.272 acc120=0.337
[23:52:02] step= 74100 | loss=2.6875 | acc30=0.211 | lr=1.00e-08 | 3.99h
[23:52:20] step= 74200 | loss=3.2619 | acc30=0.056 | lr=1.00e-08 | 4.00h
[23:52:25] --- CHECKPOINT (hora 4.0) ---
[23:52:25] Checkpoint salvo: checkpoint_step00074226.pt (59.0 MB)
[23:52:30] ☁ Upload HuggingFace OK: step=74226
[23:52:38] step= 74300 | loss=2.8144 | acc30=0.160 | lr=1.00e-08 | 4.00h
[23:52:56] step= 74400 | loss=2.2264 | acc30=0.253 | lr=1.00e-08 | 4.01h
[23:53:14] step= 74500 | loss=2.1796 | acc30=0.270 | lr=1.00e-08 | 4.01h
[23:53:32] step= 74600 | loss=2.2589 | acc30=0.285 | lr=1.00e-08 | 4.02h
[23:53:49] step= 74700 | loss=2.9828 | acc30=0.114 | lr=1.00e-08 | 4.02h
[23:54:07] step= 74800 | loss=2.4116 | acc30=0.230 | lr=1.00e-08 | 4.03h
[23:54:25] step= 74900 | loss=2.2266 | acc30=0.246 | lr=1.00e-08 | 4.03h
[23:54:43] step= 75000 | loss=2.3310 | acc30=0.200 | lr=1.00e-08 | 4.04h
[23:54:43] Validando...
[23:54:59] VAL: loss=2.3535 acc30=0.200 acc60=0.272 acc120=0.337
[23:55:16] step= 75100 | loss=2.2185 | acc30=0.258 | lr=1.00e-08 | 4.05h
[23:55:34] step= 75200 | loss=2.2637 | acc30=0.210 | lr=1.00e-08 | 4.05h
[23:55:52] step= 75300 | loss=2.3100 | acc30=0.238 | lr=1.00e-08 | 4.06h
[23:56:10] step= 75400 | loss=2.5150 | acc30=0.196 | lr=1.00e-08 | 4.06h
[23:56:27] step= 75500 | loss=5.1388 | acc30=0.010 | lr=1.00e-08 | 4.07h
[23:56:45] step= 75600 | loss=2.5718 | acc30=0.147 | lr=1.00e-08 | 4.07h
[23:57:03] step= 75700 | loss=2.3952 | acc30=0.175 | lr=1.00e-08 | 4.08h
[23:57:21] step= 75800 | loss=2.2884 | acc30=0.229 | lr=1.00e-08 | 4.08h
[23:57:38] step= 75900 | loss=2.2666 | acc30=0.205 | lr=1.00e-08 | 4.09h
[23:57:56] step= 76000 | loss=3.9404 | acc30=0.087 | lr=1.00e-08 | 4.09h
[23:57:56] Validando...
[23:58:12] VAL: loss=2.3534 acc30=0.200 acc60=0.272 acc120=0.337
[23:58:30] step= 76100 | loss=2.7910 | acc30=0.101 | lr=1.00e-08 | 4.10h
[23:58:48] step= 76200 | loss=2.7517 | acc30=0.138 | lr=1.00e-08 | 4.11h
[23:59:05] step= 76300 | loss=2.0907 | acc30=0.342 | lr=1.00e-08 | 4.11h
[23:59:23] step= 76400 | loss=2.4943 | acc30=0.142 | lr=1.00e-08 | 4.12h
[23:59:41] step= 76500 | loss=2.4157 | acc30=0.230 | lr=1.00e-08 | 4.12h
[23:59:59] step= 76600 | loss=2.0711 | acc30=0.381 | lr=1.00e-08 | 4.13h
[00:00:16] step= 76700 | loss=2.0987 | acc30=0.353 | lr=1.00e-08 | 4.13h
[00:00:34] step= 76800 | loss=2.1634 | acc30=0.267 | lr=1.00e-08 | 4.14h
[00:00:52] step= 76900 | loss=2.2255 | acc30=0.243 | lr=1.00e-08 | 4.14h
[00:01:10] step= 77000 | loss=2.4286 | acc30=0.161 | lr=1.00e-08 | 4.15h
[00:01:10] Validando...
[00:01:26] VAL: loss=2.3534 acc30=0.200 acc60=0.272 acc120=0.337
[00:01:43] step= 77100 | loss=2.2070 | acc30=0.264 | lr=1.00e-08 | 4.16h
[00:02:01] step= 77200 | loss=2.4859 | acc30=0.164 | lr=1.00e-08 | 4.16h
[00:02:19] step= 77300 | loss=2.5836 | acc30=0.158 | lr=1.00e-08 | 4.17h
[00:02:37] step= 77400 | loss=2.2187 | acc30=0.255 | lr=1.00e-08 | 4.17h
[00:02:54] step= 77500 | loss=2.5248 | acc30=0.160 | lr=1.00e-08 | 4.18h
[00:03:12] step= 77600 | loss=2.4179 | acc30=0.160 | lr=1.00e-08 | 4.18h
[00:03:30] step= 77700 | loss=2.4676 | acc30=0.175 | lr=1.00e-08 | 4.19h
[00:03:48] step= 77800 | loss=2.4459 | acc30=0.158 | lr=1.00e-08 | 4.19h
[00:04:05] step= 77900 | loss=2.6536 | acc30=0.126 | lr=1.00e-08 | 4.19h
[00:04:23] step= 78000 | loss=2.2557 | acc30=0.237 | lr=1.00e-08 | 4.20h
[00:04:23] Validando...
[00:04:39] VAL: loss=2.3534 acc30=0.200 acc60=0.272 acc120=0.337
[00:04:57] step= 78100 | loss=2.1254 | acc30=0.305 | lr=1.00e-08 | 4.21h
[00:05:14] step= 78200 | loss=2.1568 | acc30=0.281 | lr=1.00e-08 | 4.21h
[00:05:33] step= 78300 | loss=2.2215 | acc30=0.249 | lr=1.00e-08 | 4.22h
[00:05:50] step= 78400 | loss=2.2561 | acc30=0.217 | lr=1.00e-08 | 4.22h
[00:06:08] step= 78500 | loss=2.1517 | acc30=0.286 | lr=1.00e-08 | 4.23h
[00:06:26] step= 78600 | loss=2.0436 | acc30=0.399 | lr=1.00e-08 | 4.23h
[00:06:43] step= 78700 | loss=2.4710 | acc30=0.154 | lr=1.00e-08 | 4.24h
[00:07:02] step= 78800 | loss=2.4042 | acc30=0.191 | lr=1.00e-08 | 4.24h
[00:07:20] step= 78900 | loss=2.2397 | acc30=0.250 | lr=1.00e-08 | 4.25h
[00:07:37] step= 79000 | loss=2.1941 | acc30=0.265 | lr=1.00e-08 | 4.25h
[00:07:37] Validando...
[00:07:54] VAL: loss=2.3533 acc30=0.200 acc60=0.272 acc120=0.337
[00:08:11] step= 79100 | loss=2.2282 | acc30=0.251 | lr=1.00e-08 | 4.26h
[00:08:29] step= 79200 | loss=2.2841 | acc30=0.228 | lr=1.00e-08 | 4.27h
[00:08:47] step= 79300 | loss=2.8021 | acc30=0.207 | lr=1.00e-08 | 4.27h
[00:09:05] step= 79400 | loss=4.0657 | acc30=0.029 | lr=1.00e-08 | 4.28h
[00:09:22] step= 79500 | loss=2.3787 | acc30=0.218 | lr=1.00e-08 | 4.28h
[00:09:40] step= 79600 | loss=2.1206 | acc30=0.313 | lr=1.00e-08 | 4.29h
[00:09:58] step= 79700 | loss=2.1218 | acc30=0.311 | lr=1.00e-08 | 4.29h
[00:10:16] step= 79800 | loss=2.0991 | acc30=0.340 | lr=1.00e-08 | 4.30h
[00:10:34] step= 79900 | loss=2.1184 | acc30=0.311 | lr=1.00e-08 | 4.30h
[00:10:52] step= 80000 | loss=2.4259 | acc30=0.194 | lr=1.00e-08 | 4.31h
[00:10:52] Validando...
[00:11:07] VAL: loss=2.3533 acc30=0.200 acc60=0.271 acc120=0.337
[00:11:25] step= 80100 | loss=2.5228 | acc30=0.151 | lr=1.00e-08 | 4.32h
[00:11:43] step= 80200 | loss=2.3003 | acc30=0.233 | lr=1.00e-08 | 4.32h
[00:12:00] step= 80300 | loss=2.3775 | acc30=0.171 | lr=1.00e-08 | 4.33h
[00:12:18] step= 80400 | loss=2.5570 | acc30=0.130 | lr=1.00e-08 | 4.33h
[00:12:36] step= 80500 | loss=2.2503 | acc30=0.240 | lr=1.00e-08 | 4.34h
[00:12:53] step= 80600 | loss=2.2121 | acc30=0.259 | lr=1.00e-08 | 4.34h
[00:13:11] step= 80700 | loss=2.7572 | acc30=0.135 | lr=1.00e-08 | 4.35h
[00:13:29] step= 80800 | loss=2.5788 | acc30=0.185 | lr=1.00e-08 | 4.35h
[00:13:47] step= 80900 | loss=2.1669 | acc30=0.282 | lr=1.00e-08 | 4.36h
[00:14:04] step= 81000 | loss=2.3128 | acc30=0.215 | lr=1.00e-08 | 4.36h
[00:14:04] Validando...
[00:14:20] VAL: loss=2.3533 acc30=0.200 acc60=0.272 acc120=0.337
[00:14:38] step= 81100 | loss=2.8883 | acc30=0.098 | lr=1.00e-08 | 4.37h
[00:14:56] step= 81200 | loss=2.2638 | acc30=0.249 | lr=1.00e-08 | 4.38h
[00:15:14] step= 81300 | loss=2.1265 | acc30=0.322 | lr=1.00e-08 | 4.38h
[00:15:31] step= 81400 | loss=2.0707 | acc30=0.375 | lr=1.00e-08 | 4.39h
[00:15:49] step= 81500 | loss=2.7182 | acc30=0.141 | lr=1.00e-08 | 4.39h
[00:16:07] step= 81600 | loss=2.4895 | acc30=0.152 | lr=1.00e-08 | 4.40h
[00:16:25] step= 81700 | loss=2.4452 | acc30=0.170 | lr=1.00e-08 | 4.40h
[00:16:43] step= 81800 | loss=2.1218 | acc30=0.324 | lr=1.00e-08 | 4.41h
[00:17:01] step= 81900 | loss=2.2478 | acc30=0.246 | lr=1.00e-08 | 4.41h
[00:17:18] step= 82000 | loss=2.4908 | acc30=0.180 | lr=1.00e-08 | 4.42h
[00:17:18] Validando...
[00:17:35] VAL: loss=2.3533 acc30=0.200 acc60=0.272 acc120=0.337
[00:17:53] step= 82100 | loss=2.6056 | acc30=0.137 | lr=1.00e-08 | 4.42h
[00:18:10] step= 82200 | loss=3.3115 | acc30=0.072 | lr=1.00e-08 | 4.43h
[00:18:28] step= 82300 | loss=2.2575 | acc30=0.228 | lr=1.00e-08 | 4.43h
[00:18:46] step= 82400 | loss=2.1635 | acc30=0.292 | lr=1.00e-08 | 4.44h
[00:19:03] step= 82500 | loss=2.1539 | acc30=0.310 | lr=1.00e-08 | 4.44h
[00:19:21] step= 82600 | loss=2.1279 | acc30=0.324 | lr=1.00e-08 | 4.45h
[00:19:39] step= 82700 | loss=2.4373 | acc30=0.181 | lr=1.00e-08 | 4.45h
[00:19:57] step= 82800 | loss=2.3395 | acc30=0.195 | lr=1.00e-08 | 4.46h
[00:20:15] step= 82900 | loss=2.3386 | acc30=0.194 | lr=1.00e-08 | 4.46h
[00:20:32] step= 83000 | loss=2.5772 | acc30=0.140 | lr=1.00e-08 | 4.47h
[00:20:32] Validando...
[00:20:48] VAL: loss=2.3533 acc30=0.200 acc60=0.272 acc120=0.337
[00:21:06] step= 83100 | loss=2.2887 | acc30=0.215 | lr=1.00e-08 | 4.48h
[00:21:24] step= 83200 | loss=2.7948 | acc30=0.147 | lr=1.00e-08 | 4.48h
[00:21:42] step= 83300 | loss=2.6500 | acc30=0.138 | lr=1.00e-08 | 4.49h
[00:22:00] step= 83400 | loss=2.2567 | acc30=0.231 | lr=1.00e-08 | 4.49h
[00:22:18] step= 83500 | loss=2.7521 | acc30=0.117 | lr=1.00e-08 | 4.50h
[00:22:36] step= 83600 | loss=4.0087 | acc30=0.043 | lr=1.00e-08 | 4.50h
[00:22:54] step= 83700 | loss=4.0047 | acc30=0.059 | lr=1.00e-08 | 4.51h
[00:23:11] step= 83800 | loss=2.1873 | acc30=0.287 | lr=1.00e-08 | 4.51h
[00:23:29] step= 83900 | loss=2.5758 | acc30=0.153 | lr=1.00e-08 | 4.52h
[00:23:47] step= 84000 | loss=2.3208 | acc30=0.207 | lr=1.00e-08 | 4.52h
[00:23:47] Validando...
[00:24:03] VAL: loss=2.3532 acc30=0.200 acc60=0.272 acc120=0.337
[00:24:21] step= 84100 | loss=2.2624 | acc30=0.237 | lr=1.00e-08 | 4.53h
[00:24:38] step= 84200 | loss=2.2386 | acc30=0.232 | lr=1.00e-08 | 4.54h
[00:24:56] step= 84300 | loss=2.3093 | acc30=0.198 | lr=1.00e-08 | 4.54h
[00:25:14] step= 84400 | loss=2.3952 | acc30=0.180 | lr=1.00e-08 | 4.55h
[00:25:33] step= 84500 | loss=2.4518 | acc30=0.161 | lr=1.00e-08 | 4.55h
[00:25:50] step= 84600 | loss=2.9254 | acc30=0.104 | lr=1.00e-08 | 4.56h
[00:26:08] step= 84700 | loss=4.6048 | acc30=0.059 | lr=1.00e-08 | 4.56h
[00:26:26] step= 84800 | loss=3.2535 | acc30=0.086 | lr=1.00e-08 | 4.57h
[00:26:44] step= 84900 | loss=2.2908 | acc30=0.230 | lr=1.00e-08 | 4.57h
[00:27:01] step= 85000 | loss=2.0889 | acc30=0.348 | lr=1.00e-08 | 4.58h
[00:27:01] Validando...
[00:27:17] VAL: loss=2.3532 acc30=0.200 acc60=0.272 acc120=0.337
[00:27:35] step= 85100 | loss=2.4219 | acc30=0.199 | lr=1.00e-08 | 4.59h
[00:27:53] step= 85200 | loss=2.3893 | acc30=0.189 | lr=1.00e-08 | 4.59h
[00:28:11] step= 85300 | loss=2.6776 | acc30=0.116 | lr=1.00e-08 | 4.60h
[00:28:28] step= 85400 | loss=2.2643 | acc30=0.233 | lr=1.00e-08 | 4.60h
[00:28:46] step= 85500 | loss=2.5932 | acc30=0.148 | lr=1.00e-08 | 4.61h
[00:29:04] step= 85600 | loss=2.6078 | acc30=0.171 | lr=1.00e-08 | 4.61h
[00:29:22] step= 85700 | loss=2.1420 | acc30=0.292 | lr=1.00e-08 | 4.62h
[00:29:40] step= 85800 | loss=3.1185 | acc30=0.086 | lr=1.00e-08 | 4.62h
[00:29:58] step= 85900 | loss=2.5831 | acc30=0.193 | lr=1.00e-08 | 4.63h
[00:30:15] step= 86000 | loss=2.2165 | acc30=0.286 | lr=1.00e-08 | 4.63h
[00:30:15] Validando...
[00:30:32] VAL: loss=2.3532 acc30=0.200 acc60=0.273 acc120=0.337
[00:30:49] step= 86100 | loss=2.5826 | acc30=0.127 | lr=1.00e-08 | 4.64h
[00:31:07] step= 86200 | loss=2.4720 | acc30=0.146 | lr=1.00e-08 | 4.65h
[00:31:25] step= 86300 | loss=2.2958 | acc30=0.200 | lr=1.00e-08 | 4.65h
[00:31:43] step= 86400 | loss=2.2609 | acc30=0.207 | lr=1.00e-08 | 4.66h
[00:32:00] step= 86500 | loss=2.2499 | acc30=0.224 | lr=1.00e-08 | 4.66h
[00:32:18] step= 86600 | loss=2.1324 | acc30=0.325 | lr=1.00e-08 | 4.67h
[00:32:36] step= 86700 | loss=2.1140 | acc30=0.318 | lr=1.00e-08 | 4.67h
[00:32:54] step= 86800 | loss=2.3307 | acc30=0.235 | lr=1.00e-08 | 4.68h
[00:33:11] step= 86900 | loss=2.8455 | acc30=0.097 | lr=1.00e-08 | 4.68h
[00:33:29] step= 87000 | loss=2.1779 | acc30=0.345 | lr=1.00e-08 | 4.68h
[00:33:29] Validando...
[00:33:45] VAL: loss=2.3532 acc30=0.200 acc60=0.272 acc120=0.337
[00:34:03] step= 87100 | loss=2.8980 | acc30=0.206 | lr=1.00e-08 | 4.69h
[00:34:22] step= 87200 | loss=2.7367 | acc30=0.121 | lr=1.00e-08 | 4.70h
[00:34:39] step= 87300 | loss=2.3402 | acc30=0.213 | lr=1.00e-08 | 4.70h
[00:34:57] step= 87400 | loss=2.4707 | acc30=0.151 | lr=1.00e-08 | 4.71h
[00:35:15] step= 87500 | loss=2.2269 | acc30=0.255 | lr=1.00e-08 | 4.71h
[00:35:33] step= 87600 | loss=2.1344 | acc30=0.316 | lr=1.00e-08 | 4.72h
[00:35:50] step= 87700 | loss=2.0423 | acc30=0.405 | lr=1.00e-08 | 4.72h
[00:36:08] step= 87800 | loss=2.2508 | acc30=0.269 | lr=1.00e-08 | 4.73h
[00:36:26] step= 87900 | loss=2.3772 | acc30=0.203 | lr=1.00e-08 | 4.73h
[00:36:44] step= 88000 | loss=2.3070 | acc30=0.227 | lr=1.00e-08 | 4.74h
[00:36:44] Validando...
[00:37:00] VAL: loss=2.3531 acc30=0.200 acc60=0.272 acc120=0.337
[00:37:18] step= 88100 | loss=2.1593 | acc30=0.282 | lr=1.00e-08 | 4.75h
[00:37:36] step= 88200 | loss=2.8854 | acc30=0.112 | lr=1.00e-08 | 4.75h
[00:37:53] step= 88300 | loss=2.9207 | acc30=0.079 | lr=1.00e-08 | 4.76h
[00:38:11] step= 88400 | loss=2.5137 | acc30=0.160 | lr=1.00e-08 | 4.76h
[00:38:28] step= 88500 | loss=2.0933 | acc30=0.350 | lr=1.00e-08 | 4.77h
[00:38:46] step= 88600 | loss=2.2396 | acc30=0.240 | lr=1.00e-08 | 4.77h
[00:39:04] step= 88700 | loss=2.1767 | acc30=0.264 | lr=1.00e-08 | 4.78h
[00:39:22] step= 88800 | loss=2.1346 | acc30=0.330 | lr=1.00e-08 | 4.78h
[00:39:40] step= 88900 | loss=2.5809 | acc30=0.157 | lr=1.00e-08 | 4.79h
[00:39:57] step= 89000 | loss=2.4324 | acc30=0.174 | lr=1.00e-08 | 4.79h
[00:39:57] Validando...
[00:40:13] VAL: loss=2.3531 acc30=0.200 acc60=0.273 acc120=0.337
[00:40:31] step= 89100 | loss=2.3227 | acc30=0.209 | lr=1.00e-08 | 4.80h
[00:40:49] step= 89200 | loss=2.3196 | acc30=0.200 | lr=1.00e-08 | 4.81h
[00:41:07] step= 89300 | loss=2.2744 | acc30=0.239 | lr=1.00e-08 | 4.81h
[00:41:25] step= 89400 | loss=2.2478 | acc30=0.237 | lr=1.00e-08 | 4.82h
[00:41:42] step= 89500 | loss=2.7799 | acc30=0.197 | lr=1.00e-08 | 4.82h
[00:42:00] step= 89600 | loss=3.6156 | acc30=0.044 | lr=1.00e-08 | 4.83h
[00:42:18] step= 89700 | loss=2.2157 | acc30=0.252 | lr=1.00e-08 | 4.83h
[00:42:36] step= 89800 | loss=2.1825 | acc30=0.262 | lr=1.00e-08 | 4.84h
[00:42:54] step= 89900 | loss=2.3411 | acc30=0.234 | lr=1.00e-08 | 4.84h
[00:43:11] step= 90000 | loss=2.4767 | acc30=0.161 | lr=1.00e-08 | 4.85h
[00:43:11] Validando...
[00:43:29] VAL: loss=2.3531 acc30=0.200 acc60=0.272 acc120=0.337
[00:43:46] step= 90100 | loss=3.2848 | acc30=0.057 | lr=1.00e-08 | 4.86h
[00:44:04] step= 90200 | loss=2.6538 | acc30=0.167 | lr=1.00e-08 | 4.86h
[00:44:22] step= 90300 | loss=2.3083 | acc30=0.189 | lr=1.00e-08 | 4.87h
[00:44:39] step= 90400 | loss=2.1491 | acc30=0.288 | lr=1.00e-08 | 4.87h
[00:44:57] step= 90500 | loss=3.3686 | acc30=0.062 | lr=1.00e-08 | 4.88h
[00:45:15] step= 90600 | loss=2.8577 | acc30=0.155 | lr=1.00e-08 | 4.88h
[00:45:33] step= 90700 | loss=2.2129 | acc30=0.246 | lr=1.00e-08 | 4.89h
[00:45:50] step= 90800 | loss=2.4427 | acc30=0.163 | lr=1.00e-08 | 4.89h
[00:46:08] step= 90900 | loss=2.6242 | acc30=0.120 | lr=1.00e-08 | 4.90h
[00:46:26] step= 91000 | loss=3.2288 | acc30=0.066 | lr=1.00e-08 | 4.90h
[00:46:26] Validando...
[00:46:42] VAL: loss=2.3531 acc30=0.200 acc60=0.272 acc120=0.337
[00:47:00] step= 91100 | loss=2.5089 | acc30=0.193 | lr=1.00e-08 | 4.91h
[00:47:17] step= 91200 | loss=2.1437 | acc30=0.309 | lr=1.00e-08 | 4.91h
[00:47:36] step= 91300 | loss=2.2491 | acc30=0.250 | lr=1.00e-08 | 4.92h
[00:47:54] step= 91400 | loss=2.3814 | acc30=0.187 | lr=1.00e-08 | 4.92h
[00:48:12] step= 91500 | loss=2.2819 | acc30=0.216 | lr=1.00e-08 | 4.93h
[00:48:29] step= 91600 | loss=2.1579 | acc30=0.278 | lr=1.00e-08 | 4.93h
[00:48:47] step= 91700 | loss=2.5294 | acc30=0.157 | lr=1.00e-08 | 4.94h
[00:49:05] step= 91800 | loss=2.1774 | acc30=0.273 | lr=1.00e-08 | 4.94h
[00:49:23] step= 91900 | loss=2.0807 | acc30=0.371 | lr=1.00e-08 | 4.95h
[00:49:40] step= 92000 | loss=2.0560 | acc30=0.373 | lr=1.00e-08 | 4.95h
[00:49:40] Validando...
[00:49:56] VAL: loss=2.3531 acc30=0.200 acc60=0.272 acc120=0.337
[00:50:15] step= 92100 | loss=2.3561 | acc30=0.206 | lr=1.00e-08 | 4.96h
[00:50:33] step= 92200 | loss=2.1919 | acc30=0.281 | lr=1.00e-08 | 4.97h
[00:50:51] step= 92300 | loss=2.1605 | acc30=0.292 | lr=1.00e-08 | 4.97h
[00:51:08] step= 92400 | loss=2.9855 | acc30=0.094 | lr=1.00e-08 | 4.98h
[00:51:26] step= 92500 | loss=2.5838 | acc30=0.166 | lr=1.00e-08 | 4.98h
[00:51:44] step= 92600 | loss=2.3080 | acc30=0.208 | lr=1.00e-08 | 4.99h
[00:52:02] step= 92700 | loss=2.2502 | acc30=0.247 | lr=1.00e-08 | 4.99h
[00:52:19] step= 92800 | loss=2.7366 | acc30=0.111 | lr=1.00e-08 | 5.00h
[00:52:25] --- CHECKPOINT (hora 5.0) ---
[00:52:25] Checkpoint salvo: checkpoint_step00092831.pt (59.0 MB)
[00:52:31] ☁ Upload HuggingFace OK: step=92831
[00:52:37] step= 92900 | loss=2.2472 | acc30=0.282 | lr=1.00e-08 | 5.00h
[00:52:55] step= 93000 | loss=2.1347 | acc30=0.312 | lr=1.00e-08 | 5.01h
[00:52:55] Validando...
[00:53:12] VAL: loss=2.3531 acc30=0.200 acc60=0.272 acc120=0.337
[00:53:30] step= 93100 | loss=2.2548 | acc30=0.240 | lr=1.00e-08 | 5.02h
[00:53:47] step= 93200 | loss=2.2449 | acc30=0.251 | lr=1.00e-08 | 5.02h
[00:54:05] step= 93300 | loss=2.5569 | acc30=0.138 | lr=1.00e-08 | 5.03h
[00:54:23] step= 93400 | loss=2.5317 | acc30=0.156 | lr=1.00e-08 | 5.03h
[00:54:41] step= 93500 | loss=2.2582 | acc30=0.255 | lr=1.00e-08 | 5.04h
[00:54:58] step= 93600 | loss=2.2005 | acc30=0.262 | lr=1.00e-08 | 5.04h
[00:55:16] step= 93700 | loss=2.1878 | acc30=0.313 | lr=1.00e-08 | 5.05h
[00:55:34] step= 93800 | loss=2.6410 | acc30=0.252 | lr=1.00e-08 | 5.05h
[00:55:52] step= 93900 | loss=4.2053 | acc30=0.029 | lr=1.00e-08 | 5.06h
[00:56:09] step= 94000 | loss=2.2179 | acc30=0.262 | lr=1.00e-08 | 5.06h
[00:56:09] Validando...
[00:56:25] VAL: loss=2.3531 acc30=0.200 acc60=0.272 acc120=0.337
[00:56:44] step= 94100 | loss=2.4404 | acc30=0.165 | lr=1.00e-08 | 5.07h
[00:57:02] step= 94200 | loss=2.5320 | acc30=0.169 | lr=1.00e-08 | 5.08h
[00:57:19] step= 94300 | loss=2.7907 | acc30=0.135 | lr=1.00e-08 | 5.08h
[00:57:37] step= 94400 | loss=3.5283 | acc30=0.064 | lr=1.00e-08 | 5.09h
[00:57:55] step= 94500 | loss=2.6700 | acc30=0.221 | lr=1.00e-08 | 5.09h
[00:58:13] step= 94600 | loss=2.2230 | acc30=0.324 | lr=1.00e-08 | 5.10h
[00:58:30] step= 94700 | loss=2.5131 | acc30=0.133 | lr=1.00e-08 | 5.10h
[00:58:48] step= 94800 | loss=2.3487 | acc30=0.214 | lr=1.00e-08 | 5.11h
[00:59:05] step= 94900 | loss=2.4701 | acc30=0.190 | lr=1.00e-08 | 5.11h
[00:59:24] step= 95000 | loss=2.3998 | acc30=0.189 | lr=1.00e-08 | 5.12h
[00:59:24] Validando...
[00:59:40] VAL: loss=2.3531 acc30=0.200 acc60=0.272 acc120=0.337
[00:59:57] step= 95100 | loss=2.5552 | acc30=0.138 | lr=1.00e-08 | 5.13h
[01:00:15] step= 95200 | loss=2.5159 | acc30=0.179 | lr=1.00e-08 | 5.13h
[01:00:33] step= 95300 | loss=2.3699 | acc30=0.184 | lr=1.00e-08 | 5.14h
[01:00:50] step= 95400 | loss=2.3635 | acc30=0.191 | lr=1.00e-08 | 5.14h
[01:01:08] step= 95500 | loss=2.4164 | acc30=0.177 | lr=1.00e-08 | 5.15h
[01:01:26] step= 95600 | loss=2.5358 | acc30=0.125 | lr=1.00e-08 | 5.15h
[01:01:44] step= 95700 | loss=2.5215 | acc30=0.165 | lr=1.00e-08 | 5.16h
[01:02:02] step= 95800 | loss=2.1434 | acc30=0.314 | lr=1.00e-08 | 5.16h
[01:02:19] step= 95900 | loss=2.3172 | acc30=0.204 | lr=1.00e-08 | 5.17h
[01:02:37] step= 96000 | loss=2.8138 | acc30=0.100 | lr=1.00e-08 | 5.17h
[01:02:37] Validando...
[01:02:53] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:03:11] step= 96100 | loss=2.9778 | acc30=0.081 | lr=1.00e-08 | 5.18h
[01:03:29] step= 96200 | loss=3.1225 | acc30=0.058 | lr=1.00e-08 | 5.18h
[01:03:46] step= 96300 | loss=3.1205 | acc30=0.075 | lr=1.00e-08 | 5.19h
[01:04:04] step= 96400 | loss=2.7098 | acc30=0.140 | lr=1.00e-08 | 5.19h
[01:04:22] step= 96500 | loss=2.2509 | acc30=0.247 | lr=1.00e-08 | 5.20h
[01:04:40] step= 96600 | loss=2.1332 | acc30=0.343 | lr=1.00e-08 | 5.20h
[01:04:58] step= 96700 | loss=2.7914 | acc30=0.083 | lr=1.00e-08 | 5.21h
[01:05:16] step= 96800 | loss=2.4965 | acc30=0.152 | lr=1.00e-08 | 5.21h
[01:05:34] step= 96900 | loss=2.3425 | acc30=0.204 | lr=1.00e-08 | 5.22h
[01:05:52] step= 97000 | loss=2.4699 | acc30=0.182 | lr=1.00e-08 | 5.22h
[01:05:52] Validando...
[01:06:08] VAL: loss=2.3530 acc30=0.200 acc60=0.273 acc120=0.337
[01:06:25] step= 97100 | loss=4.2788 | acc30=0.015 | lr=1.00e-08 | 5.23h
[01:06:43] step= 97200 | loss=3.1554 | acc30=0.116 | lr=1.00e-08 | 5.24h
[01:07:01] step= 97300 | loss=2.4266 | acc30=0.178 | lr=1.00e-08 | 5.24h
[01:07:19] step= 97400 | loss=2.4618 | acc30=0.188 | lr=1.00e-08 | 5.25h
[01:07:36] step= 97500 | loss=2.3304 | acc30=0.206 | lr=1.00e-08 | 5.25h
[01:07:54] step= 97600 | loss=2.1315 | acc30=0.343 | lr=1.00e-08 | 5.26h
[01:08:12] step= 97700 | loss=2.3064 | acc30=0.306 | lr=1.00e-08 | 5.26h
[01:08:30] step= 97800 | loss=2.8355 | acc30=0.094 | lr=1.00e-08 | 5.27h
[01:08:48] step= 97900 | loss=2.0648 | acc30=0.399 | lr=1.00e-08 | 5.27h
[01:09:05] step= 98000 | loss=2.0863 | acc30=0.350 | lr=1.00e-08 | 5.28h
[01:09:05] Validando...
[01:09:21] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:09:39] step= 98100 | loss=2.1473 | acc30=0.305 | lr=1.00e-08 | 5.29h
[01:09:57] step= 98200 | loss=2.3889 | acc30=0.187 | lr=1.00e-08 | 5.29h
[01:10:14] step= 98300 | loss=2.0733 | acc30=0.366 | lr=1.00e-08 | 5.30h
[01:10:33] step= 98400 | loss=2.3436 | acc30=0.228 | lr=1.00e-08 | 5.30h
[01:10:51] step= 98500 | loss=3.2794 | acc30=0.100 | lr=1.00e-08 | 5.31h
[01:11:09] step= 98600 | loss=4.0116 | acc30=0.052 | lr=1.00e-08 | 5.31h
[01:11:26] step= 98700 | loss=2.1189 | acc30=0.332 | lr=1.00e-08 | 5.32h
[01:11:44] step= 98800 | loss=2.4133 | acc30=0.212 | lr=1.00e-08 | 5.32h
[01:12:02] step= 98900 | loss=2.2093 | acc30=0.255 | lr=1.00e-08 | 5.33h
[01:12:20] step= 99000 | loss=2.1290 | acc30=0.317 | lr=1.00e-08 | 5.33h
[01:12:20] Validando...
[01:12:36] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:12:53] step= 99100 | loss=3.5198 | acc30=0.052 | lr=1.00e-08 | 5.34h
[01:13:11] step= 99200 | loss=3.1105 | acc30=0.138 | lr=1.00e-08 | 5.35h
[01:13:29] step= 99300 | loss=2.4018 | acc30=0.191 | lr=1.00e-08 | 5.35h
[01:13:46] step= 99400 | loss=2.4488 | acc30=0.171 | lr=1.00e-08 | 5.36h
[01:14:04] step= 99500 | loss=2.4056 | acc30=0.190 | lr=1.00e-08 | 5.36h
[01:14:22] step= 99600 | loss=2.5148 | acc30=0.157 | lr=1.00e-08 | 5.37h
[01:14:40] step= 99700 | loss=2.3779 | acc30=0.214 | lr=1.00e-08 | 5.37h
[01:14:57] step= 99800 | loss=2.0593 | acc30=0.440 | lr=1.00e-08 | 5.38h
[01:15:16] step= 99900 | loss=2.3570 | acc30=0.195 | lr=1.00e-08 | 5.38h
[01:15:33] step=100000 | loss=2.9071 | acc30=0.122 | lr=1.00e-08 | 5.39h
[01:15:33] Validando...
[01:15:50] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:16:08] step=100100 | loss=2.6145 | acc30=0.184 | lr=1.00e-08 | 5.40h
[01:16:25] step=100200 | loss=2.0429 | acc30=0.392 | lr=1.00e-08 | 5.40h
[01:16:43] step=100300 | loss=2.2929 | acc30=0.221 | lr=1.00e-08 | 5.41h
[01:17:01] step=100400 | loss=2.6104 | acc30=0.136 | lr=1.00e-08 | 5.41h
[01:17:19] step=100500 | loss=2.3034 | acc30=0.210 | lr=1.00e-08 | 5.42h
[01:17:36] step=100600 | loss=2.2386 | acc30=0.249 | lr=1.00e-08 | 5.42h
[01:17:54] step=100700 | loss=2.6858 | acc30=0.136 | lr=1.00e-08 | 5.43h
[01:18:13] step=100800 | loss=3.0427 | acc30=0.084 | lr=1.00e-08 | 5.43h
[01:18:31] step=100900 | loss=2.8955 | acc30=0.121 | lr=1.00e-08 | 5.44h
[01:18:48] step=101000 | loss=2.2091 | acc30=0.251 | lr=1.00e-08 | 5.44h
[01:18:48] Validando...
[01:19:04] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:19:22] step=101100 | loss=2.4277 | acc30=0.166 | lr=1.00e-08 | 5.45h
[01:19:40] step=101200 | loss=2.1826 | acc30=0.255 | lr=1.00e-08 | 5.45h
[01:19:58] step=101300 | loss=2.1567 | acc30=0.285 | lr=1.00e-08 | 5.46h
[01:20:15] step=101400 | loss=2.7252 | acc30=0.138 | lr=1.00e-08 | 5.46h
[01:20:33] step=101500 | loss=2.1870 | acc30=0.293 | lr=1.00e-08 | 5.47h
[01:20:51] step=101600 | loss=2.8291 | acc30=0.219 | lr=1.00e-08 | 5.47h
[01:21:09] step=101700 | loss=3.5852 | acc30=0.075 | lr=1.00e-08 | 5.48h
[01:21:27] step=101800 | loss=2.3296 | acc30=0.209 | lr=1.00e-08 | 5.48h
[01:21:44] step=101900 | loss=2.1553 | acc30=0.282 | lr=1.00e-08 | 5.49h
[01:22:03] step=102000 | loss=3.2468 | acc30=0.157 | lr=1.00e-08 | 5.49h
[01:22:03] Validando...
[01:22:19] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:22:37] step=102100 | loss=3.5792 | acc30=0.046 | lr=1.00e-08 | 5.50h
[01:22:54] step=102200 | loss=2.2625 | acc30=0.233 | lr=1.00e-08 | 5.51h
[01:23:12] step=102300 | loss=2.1912 | acc30=0.286 | lr=1.00e-08 | 5.51h
[01:23:30] step=102400 | loss=2.1486 | acc30=0.320 | lr=1.00e-08 | 5.52h
[01:23:48] step=102500 | loss=2.3221 | acc30=0.223 | lr=1.00e-08 | 5.52h
[01:24:05] step=102600 | loss=2.0295 | acc30=0.422 | lr=1.00e-08 | 5.53h
[01:24:23] step=102700 | loss=2.5937 | acc30=0.171 | lr=1.00e-08 | 5.53h
[01:24:41] step=102800 | loss=2.3085 | acc30=0.234 | lr=1.00e-08 | 5.54h
[01:24:59] step=102900 | loss=2.1258 | acc30=0.318 | lr=1.00e-08 | 5.54h
[01:25:16] step=103000 | loss=2.0868 | acc30=0.345 | lr=1.00e-08 | 5.55h
[01:25:16] Validando...
[01:25:32] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:25:50] step=103100 | loss=2.3368 | acc30=0.186 | lr=1.00e-08 | 5.56h
[01:26:08] step=103200 | loss=2.6206 | acc30=0.113 | lr=1.00e-08 | 5.56h
[01:26:26] step=103300 | loss=2.5886 | acc30=0.121 | lr=1.00e-08 | 5.57h
[01:26:43] step=103400 | loss=2.2239 | acc30=0.222 | lr=1.00e-08 | 5.57h
[01:27:01] step=103500 | loss=2.3493 | acc30=0.169 | lr=1.00e-08 | 5.58h
[01:27:19] step=103600 | loss=2.1841 | acc30=0.251 | lr=1.00e-08 | 5.58h
[01:27:37] step=103700 | loss=2.1406 | acc30=0.320 | lr=1.00e-08 | 5.59h
[01:27:55] step=103800 | loss=2.1703 | acc30=0.284 | lr=1.00e-08 | 5.59h
[01:28:13] step=103900 | loss=2.3670 | acc30=0.194 | lr=1.00e-08 | 5.60h
[01:28:31] step=104000 | loss=2.1452 | acc30=0.296 | lr=1.00e-08 | 5.60h
[01:28:31] Validando...
[01:28:47] VAL: loss=2.3530 acc30=0.200 acc60=0.272 acc120=0.337
[01:29:04] step=104100 | loss=2.1407 | acc30=0.315 | lr=1.00e-08 | 5.61h
[01:29:22] step=104200 | loss=2.1540 | acc30=0.290 | lr=1.00e-08 | 5.62h
[01:29:40] step=104300 | loss=2.0898 | acc30=0.361 | lr=1.00e-08 | 5.62h
[01:29:58] step=104400 | loss=2.1502 | acc30=0.363 | lr=1.00e-08 | 5.63h
[01:30:16] step=104500 | loss=2.1590 | acc30=0.287 | lr=1.00e-08 | 5.63h
[01:30:33] step=104600 | loss=2.0673 | acc30=0.428 | lr=1.00e-08 | 5.64h
[01:30:51] step=104700 | loss=2.0670 | acc30=0.391 | lr=1.00e-08 | 5.64h
[01:31:09] step=104800 | loss=2.2540 | acc30=0.301 | lr=1.00e-08 | 5.65h
[01:31:26] step=104900 | loss=2.8976 | acc30=0.091 | lr=1.00e-08 | 5.65h
[01:31:44] step=105000 | loss=3.4393 | acc30=0.043 | lr=1.00e-08 | 5.66h
[01:31:44] Validando...
[01:32:00] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:32:18] step=105100 | loss=3.1834 | acc30=0.061 | lr=1.00e-08 | 5.67h
[01:32:36] step=105200 | loss=2.8606 | acc30=0.112 | lr=1.00e-08 | 5.67h
[01:32:53] step=105300 | loss=3.2665 | acc30=0.067 | lr=1.00e-08 | 5.67h
[01:33:11] step=105400 | loss=2.4372 | acc30=0.190 | lr=1.00e-08 | 5.68h
[01:33:29] step=105500 | loss=2.3841 | acc30=0.186 | lr=1.00e-08 | 5.68h
[01:33:47] step=105600 | loss=2.1814 | acc30=0.275 | lr=1.00e-08 | 5.69h
[01:34:04] step=105700 | loss=2.3208 | acc30=0.187 | lr=1.00e-08 | 5.69h
[01:34:23] step=105800 | loss=2.7356 | acc30=0.098 | lr=1.00e-08 | 5.70h
[01:34:40] step=105900 | loss=2.7699 | acc30=0.108 | lr=1.00e-08 | 5.70h
[01:34:58] step=106000 | loss=2.4355 | acc30=0.167 | lr=1.00e-08 | 5.71h
[01:34:58] Validando...
[01:35:14] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:35:32] step=106100 | loss=2.2479 | acc30=0.227 | lr=1.00e-08 | 5.72h
[01:35:50] step=106200 | loss=2.5584 | acc30=0.178 | lr=1.00e-08 | 5.72h
[01:36:08] step=106300 | loss=2.2865 | acc30=0.241 | lr=1.00e-08 | 5.73h
[01:36:25] step=106400 | loss=2.1443 | acc30=0.306 | lr=1.00e-08 | 5.73h
[01:36:43] step=106500 | loss=2.1909 | acc30=0.263 | lr=1.00e-08 | 5.74h
[01:37:00] step=106600 | loss=2.5099 | acc30=0.164 | lr=1.00e-08 | 5.74h
[01:37:19] step=106700 | loss=2.1939 | acc30=0.266 | lr=1.00e-08 | 5.75h
[01:37:37] step=106800 | loss=2.1301 | acc30=0.314 | lr=1.00e-08 | 5.75h
[01:37:54] step=106900 | loss=2.0788 | acc30=0.350 | lr=1.00e-08 | 5.76h
[01:38:12] step=107000 | loss=2.3671 | acc30=0.202 | lr=1.00e-08 | 5.76h
[01:38:12] Validando...
[01:38:28] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:38:46] step=107100 | loss=2.3523 | acc30=0.190 | lr=1.00e-08 | 5.77h
[01:39:04] step=107200 | loss=2.2676 | acc30=0.233 | lr=1.00e-08 | 5.78h
[01:39:21] step=107300 | loss=2.0970 | acc30=0.346 | lr=1.00e-08 | 5.78h
[01:39:39] step=107400 | loss=2.9710 | acc30=0.116 | lr=1.00e-08 | 5.79h
[01:39:58] step=107500 | loss=2.7002 | acc30=0.125 | lr=1.00e-08 | 5.79h
[01:40:16] step=107600 | loss=2.3089 | acc30=0.198 | lr=1.00e-08 | 5.80h
[01:40:33] step=107700 | loss=2.4801 | acc30=0.141 | lr=1.00e-08 | 5.80h
[01:40:51] step=107800 | loss=2.2628 | acc30=0.236 | lr=1.00e-08 | 5.81h
[01:41:09] step=107900 | loss=2.1562 | acc30=0.279 | lr=1.00e-08 | 5.81h
[01:41:27] step=108000 | loss=2.1482 | acc30=0.309 | lr=1.00e-08 | 5.82h
[01:41:27] Validando...
[01:41:43] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:42:00] step=108100 | loss=2.0897 | acc30=0.335 | lr=1.00e-08 | 5.83h
[01:42:18] step=108200 | loss=2.1637 | acc30=0.287 | lr=1.00e-08 | 5.83h
[01:42:36] step=108300 | loss=2.2181 | acc30=0.263 | lr=1.00e-08 | 5.84h
[01:42:54] step=108400 | loss=2.2600 | acc30=0.249 | lr=1.00e-08 | 5.84h
[01:43:11] step=108500 | loss=2.7058 | acc30=0.112 | lr=1.00e-08 | 5.85h
[01:43:29] step=108600 | loss=2.3907 | acc30=0.183 | lr=1.00e-08 | 5.85h
[01:43:47] step=108700 | loss=2.4806 | acc30=0.157 | lr=1.00e-08 | 5.86h
[01:44:05] step=108800 | loss=2.5159 | acc30=0.150 | lr=1.00e-08 | 5.86h
[01:44:23] step=108900 | loss=2.2085 | acc30=0.266 | lr=1.00e-08 | 5.87h
[01:44:41] step=109000 | loss=2.2522 | acc30=0.237 | lr=1.00e-08 | 5.87h
[01:44:41] Validando...
[01:44:57] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:45:15] step=109100 | loss=2.2587 | acc30=0.234 | lr=1.00e-08 | 5.88h
[01:45:32] step=109200 | loss=2.3955 | acc30=0.164 | lr=1.00e-08 | 5.89h
[01:45:50] step=109300 | loss=2.1208 | acc30=0.328 | lr=1.00e-08 | 5.89h
[01:46:08] step=109400 | loss=2.1072 | acc30=0.324 | lr=1.00e-08 | 5.90h
[01:46:26] step=109500 | loss=2.0982 | acc30=0.388 | lr=1.00e-08 | 5.90h
[01:46:43] step=109600 | loss=2.2031 | acc30=0.261 | lr=1.00e-08 | 5.91h
[01:47:01] step=109700 | loss=2.1784 | acc30=0.268 | lr=1.00e-08 | 5.91h
[01:47:19] step=109800 | loss=2.2108 | acc30=0.266 | lr=1.00e-08 | 5.92h
[01:47:37] step=109900 | loss=2.1707 | acc30=0.296 | lr=1.00e-08 | 5.92h
[01:47:55] step=110000 | loss=2.6104 | acc30=0.139 | lr=1.00e-08 | 5.93h
[01:47:55] Validando...
[01:48:11] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:48:29] step=110100 | loss=2.2326 | acc30=0.228 | lr=1.00e-08 | 5.93h
[01:48:47] step=110200 | loss=2.6843 | acc30=0.121 | lr=1.00e-08 | 5.94h
[01:49:05] step=110300 | loss=2.4325 | acc30=0.159 | lr=1.00e-08 | 5.94h
[01:49:23] step=110400 | loss=2.3673 | acc30=0.185 | lr=1.00e-08 | 5.95h
[01:49:41] step=110500 | loss=2.6329 | acc30=0.121 | lr=1.00e-08 | 5.95h
[01:49:59] step=110600 | loss=2.3148 | acc30=0.208 | lr=1.00e-08 | 5.96h
[01:50:16] step=110700 | loss=2.1764 | acc30=0.283 | lr=1.00e-08 | 5.96h
[01:50:34] step=110800 | loss=2.0578 | acc30=0.421 | lr=1.00e-08 | 5.97h
[01:50:52] step=110900 | loss=2.1502 | acc30=0.314 | lr=1.00e-08 | 5.97h
[01:51:10] step=111000 | loss=2.1723 | acc30=0.303 | lr=1.00e-08 | 5.98h
[01:51:10] Validando...
[01:51:26] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:51:44] step=111100 | loss=2.2418 | acc30=0.247 | lr=1.00e-08 | 5.99h
[01:52:01] step=111200 | loss=2.6688 | acc30=0.127 | lr=1.00e-08 | 5.99h
[01:52:19] step=111300 | loss=2.3549 | acc30=0.208 | lr=1.00e-08 | 6.00h
[01:52:25] --- CHECKPOINT (hora 6.0) ---
[01:52:25] Checkpoint salvo: checkpoint_step00111336.pt (59.0 MB)
[01:52:32] ☁ Upload HuggingFace OK: step=111336
[01:52:38] step=111400 | loss=2.1353 | acc30=0.313 | lr=1.00e-08 | 6.00h
[01:52:56] step=111500 | loss=2.0773 | acc30=0.363 | lr=1.00e-08 | 6.01h
[01:53:13] step=111600 | loss=2.4694 | acc30=0.144 | lr=1.00e-08 | 6.01h
[01:53:31] step=111700 | loss=2.6009 | acc30=0.116 | lr=1.00e-08 | 6.02h
[01:53:49] step=111800 | loss=2.4973 | acc30=0.182 | lr=1.00e-08 | 6.02h
[01:54:07] step=111900 | loss=2.1744 | acc30=0.267 | lr=1.00e-08 | 6.03h
[01:54:24] step=112000 | loss=2.1261 | acc30=0.325 | lr=1.00e-08 | 6.03h
[01:54:24] Validando...
[01:54:40] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:54:58] step=112100 | loss=2.2413 | acc30=0.257 | lr=1.00e-08 | 6.04h
[01:55:16] step=112200 | loss=2.0860 | acc30=0.340 | lr=1.00e-08 | 6.05h
[01:55:34] step=112300 | loss=2.1003 | acc30=0.352 | lr=1.00e-08 | 6.05h
[01:55:52] step=112400 | loss=2.2868 | acc30=0.230 | lr=1.00e-08 | 6.06h
[01:56:09] step=112500 | loss=2.2805 | acc30=0.210 | lr=1.00e-08 | 6.06h
[01:56:27] step=112600 | loss=2.3902 | acc30=0.195 | lr=1.00e-08 | 6.07h
[01:56:45] step=112700 | loss=2.4570 | acc30=0.163 | lr=1.00e-08 | 6.07h
[01:57:03] step=112800 | loss=2.4908 | acc30=0.188 | lr=1.00e-08 | 6.08h
[01:57:21] step=112900 | loss=3.3300 | acc30=0.099 | lr=1.00e-08 | 6.08h
[01:57:39] step=113000 | loss=2.3450 | acc30=0.225 | lr=1.00e-08 | 6.09h
[01:57:39] Validando...
[01:57:55] VAL: loss=2.3529 acc30=0.200 acc60=0.272 acc120=0.337
[01:58:12] step=113100 | loss=2.0822 | acc30=0.345 | lr=1.00e-08 | 6.10h
[01:58:30] step=113200 | loss=2.1486 | acc30=0.301 | lr=1.00e-08 | 6.10h
[01:58:48] step=113300 | loss=2.2262 | acc30=0.253 | lr=1.00e-08 | 6.11h
[01:59:06] step=113400 | loss=2.2574 | acc30=0.224 | lr=1.00e-08 | 6.11h
[01:59:24] step=113500 | loss=2.1818 | acc30=0.264 | lr=1.00e-08 | 6.12h
[01:59:41] step=113600 | loss=2.7845 | acc30=0.097 | lr=1.00e-08 | 6.12h
[01:59:59] step=113700 | loss=3.0627 | acc30=0.083 | lr=1.00e-08 | 6.13h
[02:00:17] step=113800 | loss=2.7396 | acc30=0.108 | lr=1.00e-08 | 6.13h
[02:00:34] step=113900 | loss=2.9172 | acc30=0.088 | lr=1.00e-08 | 6.14h
[02:00:53] step=114000 | loss=2.3761 | acc30=0.204 | lr=1.00e-08 | 6.14h
[02:00:53] Validando...
[02:01:09] VAL: loss=2.3528 acc30=0.200 acc60=0.272 acc120=0.337
[02:01:27] step=114100 | loss=2.2283 | acc30=0.268 | lr=1.00e-08 | 6.15h
[02:01:45] step=114200 | loss=2.1682 | acc30=0.297 | lr=1.00e-08 | 6.16h
[02:02:02] step=114300 | loss=2.3038 | acc30=0.204 | lr=1.00e-08 | 6.16h
[02:02:20] step=114400 | loss=2.4785 | acc30=0.144 | lr=1.00e-08 | 6.17h
[02:02:38] step=114500 | loss=2.3393 | acc30=0.198 | lr=1.00e-08 | 6.17h
[02:02:56] step=114600 | loss=2.3224 | acc30=0.203 | lr=1.00e-08 | 6.18h
[02:03:13] step=114700 | loss=2.3151 | acc30=0.214 | lr=1.00e-08 | 6.18h
[02:03:31] step=114800 | loss=2.3549 | acc30=0.208 | lr=1.00e-08 | 6.19h
[02:03:49] step=114900 | loss=2.3443 | acc30=0.202 | lr=1.00e-08 | 6.19h
[02:04:07] step=115000 | loss=2.2672 | acc30=0.232 | lr=1.00e-08 | 6.20h
[02:04:07] Validando...
[02:04:23] VAL: loss=2.3528 acc30=0.200 acc60=0.272 acc120=0.337
[02:04:41] step=115100 | loss=2.1488 | acc30=0.303 | lr=1.00e-08 | 6.20h
[02:04:59] step=115200 | loss=2.3227 | acc30=0.213 | lr=1.00e-08 | 6.21h
[02:05:17] step=115300 | loss=2.3348 | acc30=0.204 | lr=1.00e-08 | 6.21h
[02:05:35] step=115400 | loss=2.1398 | acc30=0.317 | lr=1.00e-08 | 6.22h
[02:05:52] step=115500 | loss=2.5617 | acc30=0.134 | lr=1.00e-08 | 6.22h
[02:06:10] step=115600 | loss=2.2872 | acc30=0.225 | lr=1.00e-08 | 6.23h
[02:06:28] step=115700 | loss=2.6814 | acc30=0.188 | lr=1.00e-08 | 6.23h
[02:06:46] step=115800 | loss=2.4906 | acc30=0.180 | lr=1.00e-08 | 6.24h
[02:07:03] step=115900 | loss=2.0793 | acc30=0.368 | lr=1.00e-08 | 6.24h
[02:07:21] step=116000 | loss=2.3327 | acc30=0.212 | lr=1.00e-08 | 6.25h
[02:07:21] Validando...
[02:07:37] VAL: loss=2.3528 acc30=0.200 acc60=0.272 acc120=0.337
[02:07:55] step=116100 | loss=2.4035 | acc30=0.184 | lr=1.00e-08 | 6.26h
[02:08:13] step=116200 | loss=2.6843 | acc30=0.137 | lr=1.00e-08 | 6.26h
[02:08:30] step=116300 | loss=2.3426 | acc30=0.194 | lr=1.00e-08 | 6.27h
[02:08:48] step=116400 | loss=2.6590 | acc30=0.115 | lr=1.00e-08 | 6.27h
[02:09:06] step=116500 | loss=2.2724 | acc30=0.215 | lr=1.00e-08 | 6.28h
[02:09:24] step=116600 | loss=2.2085 | acc30=0.269 | lr=1.00e-08 | 6.28h
[02:09:42] step=116700 | loss=2.7634 | acc30=0.115 | lr=1.00e-08 | 6.29h
[02:10:00] step=116800 | loss=2.7272 | acc30=0.107 | lr=1.00e-08 | 6.29h
[02:10:18] step=116900 | loss=2.5975 | acc30=0.139 | lr=1.00e-08 | 6.30h
[02:10:36] step=117000 | loss=2.2465 | acc30=0.223 | lr=1.00e-08 | 6.30h
[02:10:36] Validando...
[02:10:52] VAL: loss=2.3528 acc30=0.200 acc60=0.271 acc120=0.337
[02:11:09] step=117100 | loss=2.3541 | acc30=0.197 | lr=1.00e-08 | 6.31h
[02:11:27] step=117200 | loss=2.1117 | acc30=0.346 | lr=1.00e-08 | 6.32h
[02:11:45] step=117300 | loss=2.0882 | acc30=0.366 | lr=1.00e-08 | 6.32h
[02:12:02] step=117400 | loss=2.1721 | acc30=0.279 | lr=1.00e-08 | 6.33h
[02:12:20] step=117500 | loss=2.6204 | acc30=0.139 | lr=1.00e-08 | 6.33h
[02:12:38] step=117600 | loss=2.4958 | acc30=0.134 | lr=1.00e-08 | 6.34h
[02:12:56] step=117700 | loss=2.7025 | acc30=0.134 | lr=1.00e-08 | 6.34h
[02:13:13] step=117800 | loss=3.1755 | acc30=0.082 | lr=1.00e-08 | 6.35h
[02:13:32] step=117900 | loss=2.2968 | acc30=0.233 | lr=1.00e-08 | 6.35h
[02:13:50] step=118000 | loss=2.3275 | acc30=0.192 | lr=1.00e-08 | 6.36h
[02:13:50] Validando...
[02:14:06] VAL: loss=2.3528 acc30=0.200 acc60=0.272 acc120=0.337
[02:14:23] step=118100 | loss=2.2486 | acc30=0.241 | lr=1.00e-08 | 6.37h
[02:14:41] step=118200 | loss=2.0320 | acc30=0.414 | lr=1.00e-08 | 6.37h
[02:14:59] step=118300 | loss=2.1856 | acc30=0.285 | lr=1.00e-08 | 6.38h
[02:15:17] step=118400 | loss=2.5518 | acc30=0.167 | lr=1.00e-08 | 6.38h
[02:15:35] step=118500 | loss=2.6645 | acc30=0.123 | lr=1.00e-08 | 6.39h
[02:15:52] step=118600 | loss=2.3872 | acc30=0.187 | lr=1.00e-08 | 6.39h
[02:16:10] step=118700 | loss=4.8211 | acc30=0.025 | lr=1.00e-08 | 6.40h
[02:16:28] step=118800 | loss=3.6674 | acc30=0.196 | lr=1.00e-08 | 6.40h
[02:16:46] step=118900 | loss=2.0856 | acc30=0.347 | lr=1.00e-08 | 6.41h
[02:17:03] step=119000 | loss=2.3386 | acc30=0.193 | lr=1.00e-08 | 6.41h
[02:17:03] Validando...
[02:17:19] VAL: loss=2.3528 acc30=0.200 acc60=0.271 acc120=0.337
[02:17:37] step=119100 | loss=3.0719 | acc30=0.091 | lr=1.00e-08 | 6.42h
[02:17:56] step=119200 | loss=2.6826 | acc30=0.142 | lr=1.00e-08 | 6.43h
[02:18:13] step=119300 | loss=2.2601 | acc30=0.225 | lr=1.00e-08 | 6.43h
[02:18:31] step=119400 | loss=2.2475 | acc30=0.213 | lr=1.00e-08 | 6.44h
[02:18:49] step=119500 | loss=2.3719 | acc30=0.186 | lr=1.00e-08 | 6.44h
[02:19:07] step=119600 | loss=3.3053 | acc30=0.090 | lr=1.00e-08 | 6.45h
[02:19:25] step=119700 | loss=2.6058 | acc30=0.175 | lr=1.00e-08 | 6.45h
[02:19:43] step=119800 | loss=2.2794 | acc30=0.233 | lr=1.00e-08 | 6.46h
[02:20:00] step=119900 | loss=2.5036 | acc30=0.151 | lr=1.00e-08 | 6.46h
[02:20:18] step=120000 | loss=2.6130 | acc30=0.160 | lr=1.00e-08 | 6.47h
[02:20:18] Validando...
[02:20:34] VAL: loss=2.3528 acc30=0.200 acc60=0.271 acc120=0.337
[02:20:52] step=120100 | loss=3.0443 | acc30=0.150 | lr=1.00e-08 | 6.47h
[02:21:10] step=120200 | loss=3.8869 | acc30=0.036 | lr=1.00e-08 | 6.48h
[02:21:27] step=120300 | loss=2.0905 | acc30=0.354 | lr=1.00e-08 | 6.48h
[02:21:45] step=120400 | loss=2.2575 | acc30=0.216 | lr=1.00e-08 | 6.49h
[02:22:03] step=120500 | loss=2.1733 | acc30=0.287 | lr=1.00e-08 | 6.49h
[02:22:20] step=120600 | loss=2.0793 | acc30=0.372 | lr=1.00e-08 | 6.50h
[02:22:38] step=120700 | loss=2.5023 | acc30=0.177 | lr=1.00e-08 | 6.50h
[02:22:56] step=120800 | loss=2.4274 | acc30=0.180 | lr=1.00e-08 | 6.51h
[02:23:14] step=120900 | loss=2.3404 | acc30=0.194 | lr=1.00e-08 | 6.51h
[02:23:32] step=121000 | loss=2.3542 | acc30=0.181 | lr=1.00e-08 | 6.52h
[02:23:32] Validando...
[02:23:49] VAL: loss=2.3528 acc30=0.200 acc60=0.271 acc120=0.337
[02:24:07] step=121100 | loss=2.3336 | acc30=0.206 | lr=1.00e-08 | 6.53h
[02:24:25] step=121200 | loss=2.3643 | acc30=0.186 | lr=1.00e-08 | 6.53h
[02:24:42] step=121300 | loss=2.4391 | acc30=0.166 | lr=1.00e-08 | 6.54h
[02:24:59] step=121400 | loss=2.5812 | acc30=0.129 | lr=1.00e-08 | 6.54h
[02:25:17] step=121500 | loss=2.1839 | acc30=0.301 | lr=1.00e-08 | 6.55h
[02:25:35] step=121600 | loss=2.3031 | acc30=0.211 | lr=1.00e-08 | 6.55h
[02:25:54] step=121700 | loss=2.1937 | acc30=0.271 | lr=1.00e-08 | 6.56h
[02:26:11] step=121800 | loss=2.1672 | acc30=0.279 | lr=1.00e-08 | 6.56h
[02:26:29] step=121900 | loss=2.1130 | acc30=0.333 | lr=1.00e-08 | 6.57h
[02:26:47] step=122000 | loss=2.3199 | acc30=0.211 | lr=1.00e-08 | 6.57h
[02:26:47] Validando...
[02:27:03] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:27:21] step=122100 | loss=2.5628 | acc30=0.175 | lr=1.00e-08 | 6.58h
[02:27:38] step=122200 | loss=3.3368 | acc30=0.096 | lr=1.00e-08 | 6.59h
[02:27:56] step=122300 | loss=2.8118 | acc30=0.108 | lr=1.00e-08 | 6.59h
[02:28:14] step=122400 | loss=3.2896 | acc30=0.071 | lr=1.00e-08 | 6.60h
[02:28:32] step=122500 | loss=2.8621 | acc30=0.129 | lr=1.00e-08 | 6.60h
[02:28:49] step=122600 | loss=2.2626 | acc30=0.214 | lr=1.00e-08 | 6.61h
[02:29:07] step=122700 | loss=2.3384 | acc30=0.169 | lr=1.00e-08 | 6.61h
[02:29:25] step=122800 | loss=2.1584 | acc30=0.298 | lr=1.00e-08 | 6.62h
[02:29:43] step=122900 | loss=2.0515 | acc30=0.402 | lr=1.00e-08 | 6.62h
[02:30:01] step=123000 | loss=2.0710 | acc30=0.358 | lr=1.00e-08 | 6.63h
[02:30:01] Validando...
[02:30:17] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:30:35] step=123100 | loss=2.1464 | acc30=0.287 | lr=1.00e-08 | 6.64h
[02:30:53] step=123200 | loss=2.2195 | acc30=0.246 | lr=1.00e-08 | 6.64h
[02:31:11] step=123300 | loss=2.1564 | acc30=0.298 | lr=1.00e-08 | 6.65h
[02:31:28] step=123400 | loss=2.0879 | acc30=0.343 | lr=1.00e-08 | 6.65h
[02:31:46] step=123500 | loss=2.8867 | acc30=0.131 | lr=1.00e-08 | 6.66h
[02:32:04] step=123600 | loss=2.5212 | acc30=0.156 | lr=1.00e-08 | 6.66h
[02:32:22] step=123700 | loss=2.5823 | acc30=0.127 | lr=1.00e-08 | 6.67h
[02:32:40] step=123800 | loss=2.1297 | acc30=0.319 | lr=1.00e-08 | 6.67h
[02:32:57] step=123900 | loss=2.1698 | acc30=0.281 | lr=1.00e-08 | 6.68h
[02:33:15] step=124000 | loss=2.1490 | acc30=0.293 | lr=1.00e-08 | 6.68h
[02:33:15] Validando...
[02:33:31] VAL: loss=2.3527 acc30=0.200 acc60=0.271 acc120=0.337
[02:33:49] step=124100 | loss=2.0802 | acc30=0.346 | lr=1.00e-08 | 6.69h
[02:34:06] step=124200 | loss=2.7699 | acc30=0.128 | lr=1.00e-08 | 6.70h
[02:34:24] step=124300 | loss=2.6321 | acc30=0.123 | lr=1.00e-08 | 6.70h
[02:34:42] step=124400 | loss=2.4609 | acc30=0.165 | lr=1.00e-08 | 6.71h
[02:35:00] step=124500 | loss=2.3148 | acc30=0.208 | lr=1.00e-08 | 6.71h
[02:35:18] step=124600 | loss=2.0597 | acc30=0.379 | lr=1.00e-08 | 6.71h
[02:35:36] step=124700 | loss=2.1822 | acc30=0.285 | lr=1.00e-08 | 6.72h
[02:35:54] step=124800 | loss=2.2516 | acc30=0.240 | lr=1.00e-08 | 6.72h
[02:36:11] step=124900 | loss=2.5296 | acc30=0.143 | lr=1.00e-08 | 6.73h
[02:36:29] step=125000 | loss=2.5611 | acc30=0.151 | lr=1.00e-08 | 6.73h
[02:36:29] Validando...
[02:36:45] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:37:03] step=125100 | loss=2.4710 | acc30=0.166 | lr=1.00e-08 | 6.74h
[02:37:21] step=125200 | loss=2.3363 | acc30=0.199 | lr=1.00e-08 | 6.75h
[02:37:38] step=125300 | loss=2.3882 | acc30=0.176 | lr=1.00e-08 | 6.75h
[02:37:56] step=125400 | loss=2.8032 | acc30=0.100 | lr=1.00e-08 | 6.76h
[02:38:14] step=125500 | loss=2.4033 | acc30=0.157 | lr=1.00e-08 | 6.76h
[02:38:32] step=125600 | loss=2.2647 | acc30=0.239 | lr=1.00e-08 | 6.77h
[02:38:49] step=125700 | loss=2.1816 | acc30=0.302 | lr=1.00e-08 | 6.77h
[02:39:07] step=125800 | loss=2.3522 | acc30=0.188 | lr=1.00e-08 | 6.78h
[02:39:25] step=125900 | loss=2.3884 | acc30=0.182 | lr=1.00e-08 | 6.78h
[02:39:43] step=126000 | loss=2.1581 | acc30=0.289 | lr=1.00e-08 | 6.79h
[02:39:43] Validando...
[02:39:59] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:40:16] step=126100 | loss=2.2016 | acc30=0.259 | lr=1.00e-08 | 6.80h
[02:40:35] step=126200 | loss=2.1782 | acc30=0.282 | lr=1.00e-08 | 6.80h
[02:40:53] step=126300 | loss=2.3023 | acc30=0.252 | lr=1.00e-08 | 6.81h
[02:41:10] step=126400 | loss=2.3221 | acc30=0.218 | lr=1.00e-08 | 6.81h
[02:41:28] step=126500 | loss=2.6255 | acc30=0.116 | lr=1.00e-08 | 6.82h
[02:41:46] step=126600 | loss=2.3581 | acc30=0.251 | lr=1.00e-08 | 6.82h
[02:42:04] step=126700 | loss=3.2381 | acc30=0.137 | lr=1.00e-08 | 6.83h
[02:42:22] step=126800 | loss=3.0673 | acc30=0.099 | lr=1.00e-08 | 6.83h
[02:42:39] step=126900 | loss=2.3368 | acc30=0.187 | lr=1.00e-08 | 6.84h
[02:42:57] step=127000 | loss=2.2394 | acc30=0.245 | lr=1.00e-08 | 6.84h
[02:42:57] Validando...
[02:43:13] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:43:31] step=127100 | loss=2.2453 | acc30=0.244 | lr=1.00e-08 | 6.85h
[02:43:49] step=127200 | loss=2.1933 | acc30=0.253 | lr=1.00e-08 | 6.86h
[02:44:06] step=127300 | loss=2.1737 | acc30=0.281 | lr=1.00e-08 | 6.86h
[02:44:24] step=127400 | loss=2.8310 | acc30=0.109 | lr=1.00e-08 | 6.87h
[02:44:42] step=127500 | loss=2.3737 | acc30=0.190 | lr=1.00e-08 | 6.87h
[02:45:00] step=127600 | loss=2.2653 | acc30=0.220 | lr=1.00e-08 | 6.88h
[02:45:18] step=127700 | loss=2.3196 | acc30=0.184 | lr=1.00e-08 | 6.88h
[02:45:35] step=127800 | loss=2.1668 | acc30=0.331 | lr=1.00e-08 | 6.89h
[02:45:53] step=127900 | loss=2.0757 | acc30=0.350 | lr=1.00e-08 | 6.89h
[02:46:11] step=128000 | loss=2.0807 | acc30=0.365 | lr=1.00e-08 | 6.90h
[02:46:11] Validando...
[02:46:27] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:46:45] step=128100 | loss=2.0716 | acc30=0.376 | lr=1.00e-08 | 6.91h
[02:47:03] step=128200 | loss=2.1832 | acc30=0.268 | lr=1.00e-08 | 6.91h
[02:47:21] step=128300 | loss=2.1386 | acc30=0.294 | lr=1.00e-08 | 6.92h
[02:47:38] step=128400 | loss=2.2475 | acc30=0.230 | lr=1.00e-08 | 6.92h
[02:47:56] step=128500 | loss=3.6766 | acc30=0.035 | lr=1.00e-08 | 6.93h
[02:48:14] step=128600 | loss=2.2996 | acc30=0.247 | lr=1.00e-08 | 6.93h
[02:48:32] step=128700 | loss=2.3351 | acc30=0.209 | lr=1.00e-08 | 6.94h
[02:48:49] step=128800 | loss=2.6658 | acc30=0.114 | lr=1.00e-08 | 6.94h
[02:49:07] step=128900 | loss=2.2295 | acc30=0.232 | lr=1.00e-08 | 6.95h
[02:49:25] step=129000 | loss=2.3528 | acc30=0.187 | lr=1.00e-08 | 6.95h
[02:49:25] Validando...
[02:49:41] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:49:59] step=129100 | loss=2.4138 | acc30=0.197 | lr=1.00e-08 | 6.96h
[02:50:16] step=129200 | loss=2.5689 | acc30=0.144 | lr=1.00e-08 | 6.96h
[02:50:34] step=129300 | loss=3.0998 | acc30=0.086 | lr=1.00e-08 | 6.97h
[02:50:52] step=129400 | loss=2.5162 | acc30=0.194 | lr=1.00e-08 | 6.97h
[02:51:10] step=129500 | loss=2.1674 | acc30=0.278 | lr=1.00e-08 | 6.98h
[02:51:27] step=129600 | loss=2.1925 | acc30=0.270 | lr=1.00e-08 | 6.98h
[02:51:45] step=129700 | loss=2.2147 | acc30=0.248 | lr=1.00e-08 | 6.99h
[02:52:03] step=129800 | loss=2.1812 | acc30=0.272 | lr=1.00e-08 | 6.99h
[02:52:21] step=129900 | loss=2.1162 | acc30=0.309 | lr=1.00e-08 | 7.00h
[02:52:26] --- CHECKPOINT (hora 7.0) ---
[02:52:26] Checkpoint salvo: checkpoint_step00129926.pt (59.0 MB)
[02:52:31] ☁ Upload HuggingFace OK: step=129926
[02:52:39] step=130000 | loss=2.1136 | acc30=0.312 | lr=1.00e-08 | 7.00h
[02:52:39] Validando...
[02:52:55] VAL: loss=2.3527 acc30=0.200 acc60=0.271 acc120=0.337
[02:53:13] step=130100 | loss=2.2329 | acc30=0.233 | lr=1.00e-08 | 7.01h
[02:53:31] step=130200 | loss=2.3300 | acc30=0.203 | lr=1.00e-08 | 7.02h
[02:53:49] step=130300 | loss=2.2986 | acc30=0.201 | lr=1.00e-08 | 7.02h
[02:54:07] step=130400 | loss=2.2822 | acc30=0.193 | lr=1.00e-08 | 7.03h
[02:54:25] step=130500 | loss=2.1752 | acc30=0.299 | lr=1.00e-08 | 7.03h
[02:54:43] step=130600 | loss=2.4236 | acc30=0.230 | lr=1.00e-08 | 7.04h
[02:55:01] step=130700 | loss=2.2844 | acc30=0.207 | lr=1.00e-08 | 7.04h
[02:55:18] step=130800 | loss=2.8008 | acc30=0.088 | lr=1.00e-08 | 7.05h
[02:55:36] step=130900 | loss=2.6495 | acc30=0.151 | lr=1.00e-08 | 7.05h
[02:55:54] step=131000 | loss=2.3020 | acc30=0.207 | lr=1.00e-08 | 7.06h
[02:55:54] Validando...
[02:56:10] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:56:28] step=131100 | loss=2.1771 | acc30=0.273 | lr=1.00e-08 | 7.07h
[02:56:45] step=131200 | loss=2.4803 | acc30=0.146 | lr=1.00e-08 | 7.07h
[02:57:03] step=131300 | loss=2.2913 | acc30=0.219 | lr=1.00e-08 | 7.08h
[02:57:21] step=131400 | loss=2.1591 | acc30=0.285 | lr=1.00e-08 | 7.08h
[02:57:39] step=131500 | loss=2.1691 | acc30=0.286 | lr=1.00e-08 | 7.09h
[02:57:56] step=131600 | loss=2.6936 | acc30=0.101 | lr=1.00e-08 | 7.09h
[02:58:15] step=131700 | loss=3.0350 | acc30=0.082 | lr=1.00e-08 | 7.10h
[02:58:33] step=131800 | loss=2.3120 | acc30=0.222 | lr=1.00e-08 | 7.10h
[02:58:51] step=131900 | loss=2.1520 | acc30=0.283 | lr=1.00e-08 | 7.11h
[02:59:08] step=132000 | loss=2.8846 | acc30=0.104 | lr=1.00e-08 | 7.11h
[02:59:08] Validando...
[02:59:24] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[02:59:42] step=132100 | loss=2.1744 | acc30=0.279 | lr=1.00e-08 | 7.12h
[03:00:00] step=132200 | loss=2.1565 | acc30=0.292 | lr=1.00e-08 | 7.13h
[03:00:18] step=132300 | loss=2.0997 | acc30=0.346 | lr=1.00e-08 | 7.13h
[03:00:35] step=132400 | loss=2.1682 | acc30=0.268 | lr=1.00e-08 | 7.14h
[03:00:53] step=132500 | loss=2.7262 | acc30=0.135 | lr=1.00e-08 | 7.14h
[03:01:11] step=132600 | loss=2.9303 | acc30=0.092 | lr=1.00e-08 | 7.15h
[03:01:28] step=132700 | loss=2.2655 | acc30=0.223 | lr=1.00e-08 | 7.15h
[03:01:46] step=132800 | loss=3.2887 | acc30=0.059 | lr=1.00e-08 | 7.16h
[03:02:05] step=132900 | loss=2.8384 | acc30=0.175 | lr=1.00e-08 | 7.16h
[03:02:23] step=133000 | loss=2.2207 | acc30=0.237 | lr=1.00e-08 | 7.17h
[03:02:23] Validando...
[03:02:39] VAL: loss=2.3527 acc30=0.200 acc60=0.272 acc120=0.337
[03:02:56] step=133100 | loss=2.2631 | acc30=0.222 | lr=1.00e-08 | 7.18h
[03:03:14] step=133200 | loss=2.1839 | acc30=0.266 | lr=1.00e-08 | 7.18h
[03:03:32] step=133300 | loss=3.2025 | acc30=0.134 | lr=1.00e-08 | 7.19h
[03:03:50] step=133400 | loss=4.3176 | acc30=0.027 | lr=1.00e-08 | 7.19h
[03:04:07] step=133500 | loss=5.0533 | acc30=0.012 | lr=1.00e-08 | 7.20h
[03:04:25] step=133600 | loss=2.2296 | acc30=0.276 | lr=1.00e-08 | 7.20h
[03:04:43] step=133700 | loss=2.3143 | acc30=0.207 | lr=1.00e-08 | 7.21h
[03:05:01] step=133800 | loss=2.1972 | acc30=0.260 | lr=1.00e-08 | 7.21h
[03:05:18] step=133900 | loss=2.1684 | acc30=0.277 | lr=1.00e-08 | 7.22h
[03:05:36] step=134000 | loss=2.0870 | acc30=0.343 | lr=1.00e-08 | 7.22h
[03:05:36] Validando...
[03:05:53] VAL: loss=2.3527 acc30=0.200 acc60=0.271 acc120=0.337
[03:06:11] step=134100 | loss=2.4248 | acc30=0.216 | lr=1.00e-08 | 7.23h
[03:06:29] step=134200 | loss=2.8965 | acc30=0.131 | lr=1.00e-08 | 7.23h
[03:06:46] step=134300 | loss=3.9946 | acc30=0.028 | lr=1.00e-08 | 7.24h
[03:07:04] step=134400 | loss=2.5569 | acc30=0.180 | lr=1.00e-08 | 7.24h
[03:07:22] step=134500 | loss=2.1896 | acc30=0.263 | lr=1.00e-08 | 7.25h
[03:07:40] step=134600 | loss=2.2870 | acc30=0.223 | lr=1.00e-08 | 7.25h
[03:07:57] step=134700 | loss=2.3397 | acc30=0.229 | lr=1.00e-08 | 7.26h
[03:08:15] step=134800 | loss=2.5933 | acc30=0.129 | lr=1.00e-08 | 7.26h
[03:08:33] step=134900 | loss=2.5590 | acc30=0.129 | lr=1.00e-08 | 7.27h
[03:08:51] step=135000 | loss=2.4783 | acc30=0.154 | lr=1.00e-08 | 7.27h
[03:08:51] Validando...
[03:09:07] VAL: loss=2.3526 acc30=0.200 acc60=0.272 acc120=0.337
[03:09:25] step=135100 | loss=2.1266 | acc30=0.331 | lr=1.00e-08 | 7.28h
[03:09:43] step=135200 | loss=2.1705 | acc30=0.269 | lr=1.00e-08 | 7.29h
[03:10:01] step=135300 | loss=2.4134 | acc30=0.235 | lr=1.00e-08 | 7.29h
[03:10:19] step=135400 | loss=2.9006 | acc30=0.083 | lr=1.00e-08 | 7.30h
[03:10:36] step=135500 | loss=3.3233 | acc30=0.049 | lr=1.00e-08 | 7.30h
[03:10:54] step=135600 | loss=2.6718 | acc30=0.156 | lr=1.00e-08 | 7.31h
[03:11:12] step=135700 | loss=2.2883 | acc30=0.219 | lr=1.00e-08 | 7.31h
[03:11:30] step=135800 | loss=2.3007 | acc30=0.195 | lr=1.00e-08 | 7.32h
[03:11:47] step=135900 | loss=2.3642 | acc30=0.197 | lr=1.00e-08 | 7.32h
[03:12:05] step=136000 | loss=2.3423 | acc30=0.168 | lr=1.00e-08 | 7.33h
[03:12:05] Validando...
[03:12:21] VAL: loss=2.3526 acc30=0.200 acc60=0.271 acc120=0.337
[03:12:39] step=136100 | loss=2.2966 | acc30=0.204 | lr=1.00e-08 | 7.34h
[03:12:57] step=136200 | loss=2.2367 | acc30=0.240 | lr=1.00e-08 | 7.34h
[03:13:14] step=136300 | loss=2.3623 | acc30=0.186 | lr=1.00e-08 | 7.35h
[03:13:32] step=136400 | loss=2.2291 | acc30=0.229 | lr=1.00e-08 | 7.35h
[03:13:50] step=136500 | loss=2.4861 | acc30=0.188 | lr=1.00e-08 | 7.36h
[03:14:08] step=136600 | loss=2.6070 | acc30=0.179 | lr=1.00e-08 | 7.36h
[03:14:26] step=136700 | loss=2.1062 | acc30=0.339 | lr=1.00e-08 | 7.37h
[03:14:44] step=136800 | loss=2.4801 | acc30=0.175 | lr=1.00e-08 | 7.37h
[03:15:01] step=136900 | loss=2.4092 | acc30=0.181 | lr=1.00e-08 | 7.38h
[03:15:19] step=137000 | loss=2.9591 | acc30=0.067 | lr=1.00e-08 | 7.38h
[03:15:19] Validando...
[03:15:36] VAL: loss=2.3526 acc30=0.200 acc60=0.271 acc120=0.337
[03:15:54] step=137100 | loss=2.2932 | acc30=0.214 | lr=1.00e-08 | 7.39h
[03:16:12] step=137200 | loss=2.1722 | acc30=0.293 | lr=1.00e-08 | 7.40h
[03:16:30] step=137300 | loss=2.5115 | acc30=0.246 | lr=1.00e-08 | 7.40h
[03:16:47] step=137400 | loss=3.1989 | acc30=0.065 | lr=1.00e-08 | 7.41h
[03:17:05] step=137500 | loss=2.1246 | acc30=0.383 | lr=1.00e-08 | 7.41h
[03:17:23] step=137600 | loss=2.5128 | acc30=0.216 | lr=1.00e-08 | 7.42h
[03:17:41] step=137700 | loss=2.7623 | acc30=0.129 | lr=1.00e-08 | 7.42h
[03:17:58] step=137800 | loss=2.2218 | acc30=0.253 | lr=1.00e-08 | 7.43h
[03:18:16] step=137900 | loss=2.3600 | acc30=0.179 | lr=1.00e-08 | 7.43h
[03:18:34] step=138000 | loss=2.5795 | acc30=0.132 | lr=1.00e-08 | 7.44h
[03:18:34] Validando...
[03:18:51] VAL: loss=2.3526 acc30=0.200 acc60=0.272 acc120=0.337
[03:19:09] step=138100 | loss=2.3715 | acc30=0.190 | lr=1.00e-08 | 7.45h
[03:19:26] step=138200 | loss=2.4459 | acc30=0.171 | lr=1.00e-08 | 7.45h
[03:19:44] step=138300 | loss=2.1409 | acc30=0.325 | lr=1.00e-08 | 7.46h
[03:20:02] step=138400 | loss=2.1020 | acc30=0.345 | lr=1.00e-08 | 7.46h
[03:20:20] step=138500 | loss=2.2060 | acc30=0.335 | lr=1.00e-08 | 7.47h
[03:20:37] step=138600 | loss=3.8169 | acc30=0.039 | lr=1.00e-08 | 7.47h
[03:20:55] step=138700 | loss=2.6114 | acc30=0.173 | lr=1.00e-08 | 7.48h
[03:21:13] step=138800 | loss=2.4653 | acc30=0.188 | lr=1.00e-08 | 7.48h
[03:21:31] step=138900 | loss=2.3624 | acc30=0.226 | lr=1.00e-08 | 7.49h
[03:21:48] step=139000 | loss=2.3230 | acc30=0.194 | lr=1.00e-08 | 7.49h
[03:21:48] Validando...
[03:22:04] VAL: loss=2.3526 acc30=0.200 acc60=0.272 acc120=0.337
[03:22:22] step=139100 | loss=2.3377 | acc30=0.204 | lr=1.00e-08 | 7.50h
[03:22:40] step=139200 | loss=2.2873 | acc30=0.216 | lr=1.00e-08 | 7.50h
[03:22:58] step=139300 | loss=2.5797 | acc30=0.152 | lr=1.00e-08 | 7.51h
[03:23:16] step=139400 | loss=4.1340 | acc30=0.032 | lr=1.00e-08 | 7.51h
[03:23:34] step=139500 | loss=2.4898 | acc30=0.157 | lr=1.00e-08 | 7.52h
[03:23:51] step=139600 | loss=2.4428 | acc30=0.155 | lr=1.00e-08 | 7.52h
[03:24:09] step=139700 | loss=2.4215 | acc30=0.172 | lr=1.00e-08 | 7.53h
[03:24:27] step=139800 | loss=3.0499 | acc30=0.067 | lr=1.00e-08 | 7.53h
[03:24:45] step=139900 | loss=2.2589 | acc30=0.271 | lr=1.00e-08 | 7.54h
[03:25:02] step=140000 | loss=2.2513 | acc30=0.268 | lr=1.00e-08 | 7.54h
[03:25:02] Validando...
[03:25:19] VAL: loss=2.3526 acc30=0.200 acc60=0.271 acc120=0.337
[03:25:37] step=140100 | loss=2.3543 | acc30=0.220 | lr=1.00e-08 | 7.55h
[03:25:54] step=140200 | loss=2.2226 | acc30=0.234 | lr=1.00e-08 | 7.56h
[03:26:12] step=140300 | loss=2.5463 | acc30=0.142 | lr=1.00e-08 | 7.56h
[03:26:30] step=140400 | loss=2.2195 | acc30=0.241 | lr=1.00e-08 | 7.57h
[03:26:47] step=140500 | loss=2.2081 | acc30=0.256 | lr=1.00e-08 | 7.57h
[03:27:05] step=140600 | loss=2.1359 | acc30=0.289 | lr=1.00e-08 | 7.58h
[03:27:23] step=140700 | loss=2.2398 | acc30=0.230 | lr=1.00e-08 | 7.58h
[03:27:41] step=140800 | loss=2.2455 | acc30=0.239 | lr=1.00e-08 | 7.59h
[03:27:59] step=140900 | loss=2.2401 | acc30=0.228 | lr=1.00e-08 | 7.59h
[03:28:16] step=141000 | loss=2.5632 | acc30=0.140 | lr=1.00e-08 | 7.60h
[03:28:16] Validando...
[03:28:33] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:28:51] step=141100 | loss=2.8627 | acc30=0.096 | lr=1.00e-08 | 7.61h
[03:29:09] step=141200 | loss=2.3210 | acc30=0.228 | lr=1.00e-08 | 7.61h
[03:29:26] step=141300 | loss=2.6164 | acc30=0.145 | lr=1.00e-08 | 7.62h
[03:29:44] step=141400 | loss=2.2753 | acc30=0.217 | lr=1.00e-08 | 7.62h
[03:30:02] step=141500 | loss=2.2033 | acc30=0.258 | lr=1.00e-08 | 7.63h
[03:30:20] step=141600 | loss=2.3401 | acc30=0.211 | lr=1.00e-08 | 7.63h
[03:30:37] step=141700 | loss=3.2051 | acc30=0.075 | lr=1.00e-08 | 7.64h
[03:30:55] step=141800 | loss=2.2277 | acc30=0.246 | lr=1.00e-08 | 7.64h
[03:31:13] step=141900 | loss=2.3503 | acc30=0.216 | lr=1.00e-08 | 7.65h
[03:31:31] step=142000 | loss=2.3458 | acc30=0.211 | lr=1.00e-08 | 7.65h
[03:31:31] Validando...
[03:31:47] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:32:04] step=142100 | loss=2.0790 | acc30=0.354 | lr=1.00e-08 | 7.66h
[03:32:22] step=142200 | loss=2.7107 | acc30=0.151 | lr=1.00e-08 | 7.67h
[03:32:40] step=142300 | loss=2.5720 | acc30=0.134 | lr=1.00e-08 | 7.67h
[03:32:58] step=142400 | loss=2.3249 | acc30=0.195 | lr=1.00e-08 | 7.68h
[03:33:15] step=142500 | loss=2.3020 | acc30=0.205 | lr=1.00e-08 | 7.68h
[03:33:33] step=142600 | loss=4.2631 | acc30=0.038 | lr=1.00e-08 | 7.69h
[03:33:51] step=142700 | loss=3.3204 | acc30=0.117 | lr=1.00e-08 | 7.69h
[03:34:09] step=142800 | loss=2.2218 | acc30=0.238 | lr=1.00e-08 | 7.70h
[03:34:26] step=142900 | loss=2.2991 | acc30=0.218 | lr=1.00e-08 | 7.70h
[03:34:44] step=143000 | loss=2.8548 | acc30=0.084 | lr=1.00e-08 | 7.71h
[03:34:44] Validando...
[03:35:00] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:35:18] step=143100 | loss=3.1956 | acc30=0.066 | lr=1.00e-08 | 7.72h
[03:35:36] step=143200 | loss=2.8535 | acc30=0.132 | lr=1.00e-08 | 7.72h
[03:35:53] step=143300 | loss=3.1542 | acc30=0.056 | lr=1.00e-08 | 7.72h
[03:36:11] step=143400 | loss=2.3986 | acc30=0.257 | lr=1.00e-08 | 7.73h
[03:36:29] step=143500 | loss=2.4765 | acc30=0.202 | lr=1.00e-08 | 7.73h
[03:36:47] step=143600 | loss=2.7421 | acc30=0.118 | lr=1.00e-08 | 7.74h
[03:37:04] step=143700 | loss=2.7703 | acc30=0.110 | lr=1.00e-08 | 7.74h
[03:37:22] step=143800 | loss=2.2850 | acc30=0.221 | lr=1.00e-08 | 7.75h
[03:37:41] step=143900 | loss=2.5314 | acc30=0.174 | lr=1.00e-08 | 7.75h
[03:37:59] step=144000 | loss=2.3935 | acc30=0.195 | lr=1.00e-08 | 7.76h
[03:37:59] Validando...
[03:38:15] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:38:32] step=144100 | loss=2.6091 | acc30=0.112 | lr=1.00e-08 | 7.77h
[03:38:50] step=144200 | loss=2.4279 | acc30=0.167 | lr=1.00e-08 | 7.77h
[03:39:08] step=144300 | loss=2.2207 | acc30=0.230 | lr=1.00e-08 | 7.78h
[03:39:26] step=144400 | loss=2.1224 | acc30=0.299 | lr=1.00e-08 | 7.78h
[03:39:43] step=144500 | loss=2.1377 | acc30=0.293 | lr=1.00e-08 | 7.79h
[03:40:01] step=144600 | loss=2.2406 | acc30=0.237 | lr=1.00e-08 | 7.79h
[03:40:19] step=144700 | loss=2.3844 | acc30=0.200 | lr=1.00e-08 | 7.80h
[03:40:36] step=144800 | loss=2.5968 | acc30=0.121 | lr=1.00e-08 | 7.80h
[03:40:54] step=144900 | loss=2.1304 | acc30=0.328 | lr=1.00e-08 | 7.81h
[03:41:12] step=145000 | loss=2.5667 | acc30=0.176 | lr=1.00e-08 | 7.81h
[03:41:12] Validando...
[03:41:28] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:41:46] step=145100 | loss=2.6180 | acc30=0.133 | lr=1.00e-08 | 7.82h
[03:42:03] step=145200 | loss=2.8861 | acc30=0.097 | lr=1.00e-08 | 7.83h
[03:42:21] step=145300 | loss=2.3564 | acc30=0.211 | lr=1.00e-08 | 7.83h
[03:42:39] step=145400 | loss=2.2135 | acc30=0.246 | lr=1.00e-08 | 7.84h
[03:42:57] step=145500 | loss=2.1700 | acc30=0.293 | lr=1.00e-08 | 7.84h
[03:43:14] step=145600 | loss=2.1886 | acc30=0.260 | lr=1.00e-08 | 7.85h
[03:43:32] step=145700 | loss=2.2343 | acc30=0.230 | lr=1.00e-08 | 7.85h
[03:43:50] step=145800 | loss=3.0958 | acc30=0.094 | lr=1.00e-08 | 7.86h
[03:44:08] step=145900 | loss=3.2251 | acc30=0.103 | lr=1.00e-08 | 7.86h
[03:44:25] step=146000 | loss=2.4800 | acc30=0.176 | lr=1.00e-08 | 7.87h
[03:44:25] Validando...
[03:44:41] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:44:59] step=146100 | loss=2.4689 | acc30=0.152 | lr=1.00e-08 | 7.88h
[03:45:17] step=146200 | loss=2.2178 | acc30=0.243 | lr=1.00e-08 | 7.88h
[03:45:35] step=146300 | loss=2.2900 | acc30=0.233 | lr=1.00e-08 | 7.89h
[03:45:52] step=146400 | loss=2.4120 | acc30=0.179 | lr=1.00e-08 | 7.89h
[03:46:11] step=146500 | loss=2.0873 | acc30=0.363 | lr=1.00e-08 | 7.90h
[03:46:29] step=146600 | loss=2.4949 | acc30=0.212 | lr=1.00e-08 | 7.90h
[03:46:47] step=146700 | loss=2.5899 | acc30=0.131 | lr=1.00e-08 | 7.91h
[03:47:04] step=146800 | loss=3.2041 | acc30=0.058 | lr=1.00e-08 | 7.91h
[03:47:22] step=146900 | loss=3.3101 | acc30=0.048 | lr=1.00e-08 | 7.92h
[03:47:40] step=147000 | loss=3.3900 | acc30=0.061 | lr=1.00e-08 | 7.92h
[03:47:40] Validando...
[03:47:56] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:48:14] step=147100 | loss=2.8343 | acc30=0.120 | lr=1.00e-08 | 7.93h
[03:48:31] step=147200 | loss=2.2633 | acc30=0.229 | lr=1.00e-08 | 7.94h
[03:48:49] step=147300 | loss=2.3538 | acc30=0.184 | lr=1.00e-08 | 7.94h
[03:49:07] step=147400 | loss=2.5932 | acc30=0.153 | lr=1.00e-08 | 7.95h
[03:49:25] step=147500 | loss=2.6191 | acc30=0.142 | lr=1.00e-08 | 7.95h
[03:49:42] step=147600 | loss=2.1677 | acc30=0.280 | lr=1.00e-08 | 7.96h
[03:50:00] step=147700 | loss=2.7842 | acc30=0.118 | lr=1.00e-08 | 7.96h
[03:50:18] step=147800 | loss=2.4415 | acc30=0.190 | lr=1.00e-08 | 7.97h
[03:50:36] step=147900 | loss=2.5505 | acc30=0.139 | lr=1.00e-08 | 7.97h
[03:50:54] step=148000 | loss=2.4209 | acc30=0.157 | lr=1.00e-08 | 7.98h
[03:50:54] Validando...
[03:51:10] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:51:28] step=148100 | loss=2.3079 | acc30=0.229 | lr=1.00e-08 | 7.98h
[03:51:46] step=148200 | loss=2.2305 | acc30=0.249 | lr=1.00e-08 | 7.99h
[03:52:04] step=148300 | loss=2.1528 | acc30=0.297 | lr=1.00e-08 | 7.99h
[03:52:21] step=148400 | loss=2.0574 | acc30=0.389 | lr=1.00e-08 | 8.00h
[03:52:26] --- CHECKPOINT (hora 8.0) ---
[03:52:26] Checkpoint salvo: checkpoint_step00148424.pt (59.0 MB)
[03:52:32] ☁ Upload HuggingFace OK: step=148424
[03:52:40] step=148500 | loss=2.6436 | acc30=0.138 | lr=1.00e-08 | 8.00h
[03:52:58] step=148600 | loss=2.3999 | acc30=0.178 | lr=1.00e-08 | 8.01h
[03:53:16] step=148700 | loss=2.2863 | acc30=0.231 | lr=1.00e-08 | 8.01h
[03:53:33] step=148800 | loss=2.2527 | acc30=0.218 | lr=1.00e-08 | 8.02h
[03:53:51] step=148900 | loss=2.3787 | acc30=0.179 | lr=1.00e-08 | 8.02h
[03:54:09] step=149000 | loss=2.3468 | acc30=0.212 | lr=1.00e-08 | 8.03h
[03:54:09] Validando...
[03:54:25] VAL: loss=2.3525 acc30=0.200 acc60=0.271 acc120=0.337
[03:54:43] step=149100 | loss=2.2074 | acc30=0.257 | lr=1.00e-08 | 8.04h
[03:55:00] step=149200 | loss=2.1653 | acc30=0.295 | lr=1.00e-08 | 8.04h
[03:55:18] step=149300 | loss=2.1491 | acc30=0.285 | lr=1.00e-08 | 8.05h
[03:55:36] step=149400 | loss=2.1970 | acc30=0.266 | lr=1.00e-08 | 8.05h
[03:55:54] step=149500 | loss=2.1719 | acc30=0.269 | lr=1.00e-08 | 8.06h
[03:56:11] step=149600 | loss=2.1493 | acc30=0.283 | lr=1.00e-08 | 8.06h
[03:56:29] step=149700 | loss=3.8267 | acc30=0.078 | lr=1.00e-08 | 8.07h
[03:56:47] step=149800 | loss=3.3375 | acc30=0.060 | lr=1.00e-08 | 8.07h
[03:57:05] step=149900 | loss=2.7080 | acc30=0.139 | lr=1.00e-08 | 8.08h
[03:57:23] step=150000 | loss=4.1982 | acc30=0.021 | lr=1.00e-08 | 8.08h
[03:57:23] Validando...
[03:57:39] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[03:57:56] step=150100 | loss=2.7515 | acc30=0.118 | lr=1.00e-08 | 8.09h
[03:58:14] step=150200 | loss=2.5389 | acc30=0.159 | lr=1.00e-08 | 8.10h
[03:58:32] step=150300 | loss=2.8555 | acc30=0.116 | lr=1.00e-08 | 8.10h
[03:58:49] step=150400 | loss=2.1583 | acc30=0.292 | lr=1.00e-08 | 8.11h
[03:59:07] step=150500 | loss=2.2203 | acc30=0.243 | lr=1.00e-08 | 8.11h
[03:59:25] step=150600 | loss=2.3033 | acc30=0.214 | lr=1.00e-08 | 8.12h
[03:59:43] step=150700 | loss=2.4030 | acc30=0.169 | lr=1.00e-08 | 8.12h
[04:00:01] step=150800 | loss=2.0751 | acc30=0.356 | lr=1.00e-08 | 8.13h
[04:00:19] step=150900 | loss=2.3511 | acc30=0.210 | lr=1.00e-08 | 8.13h
[04:00:37] step=151000 | loss=2.8880 | acc30=0.093 | lr=1.00e-08 | 8.14h
[04:00:37] Validando...
[04:00:53] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[04:01:10] step=151100 | loss=2.5236 | acc30=0.182 | lr=1.00e-08 | 8.15h
[04:01:28] step=151200 | loss=2.3244 | acc30=0.205 | lr=1.00e-08 | 8.15h
[04:01:46] step=151300 | loss=2.7904 | acc30=0.127 | lr=1.00e-08 | 8.16h
[04:02:04] step=151400 | loss=2.8068 | acc30=0.132 | lr=1.00e-08 | 8.16h
[04:02:21] step=151500 | loss=2.3405 | acc30=0.186 | lr=1.00e-08 | 8.17h
[04:02:39] step=151600 | loss=2.5306 | acc30=0.168 | lr=1.00e-08 | 8.17h
[04:02:57] step=151700 | loss=2.4507 | acc30=0.194 | lr=1.00e-08 | 8.18h
[04:03:15] step=151800 | loss=2.2011 | acc30=0.277 | lr=1.00e-08 | 8.18h
[04:03:33] step=151900 | loss=2.3021 | acc30=0.219 | lr=1.00e-08 | 8.19h
[04:03:51] step=152000 | loss=2.4703 | acc30=0.159 | lr=1.00e-08 | 8.19h
[04:03:51] Validando...
[04:04:07] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:04:25] step=152100 | loss=2.2401 | acc30=0.239 | lr=1.00e-08 | 8.20h
[04:04:43] step=152200 | loss=2.2152 | acc30=0.328 | lr=1.00e-08 | 8.21h
[04:05:00] step=152300 | loss=3.1544 | acc30=0.057 | lr=1.00e-08 | 8.21h
[04:05:18] step=152400 | loss=2.3403 | acc30=0.223 | lr=1.00e-08 | 8.22h
[04:05:36] step=152500 | loss=2.2645 | acc30=0.220 | lr=1.00e-08 | 8.22h
[04:05:54] step=152600 | loss=2.1304 | acc30=0.312 | lr=1.00e-08 | 8.23h
[04:06:11] step=152700 | loss=2.1084 | acc30=0.361 | lr=1.00e-08 | 8.23h
[04:06:29] step=152800 | loss=2.2703 | acc30=0.224 | lr=1.00e-08 | 8.23h
[04:06:47] step=152900 | loss=2.2738 | acc30=0.215 | lr=1.00e-08 | 8.24h
[04:07:05] step=153000 | loss=2.2264 | acc30=0.243 | lr=1.00e-08 | 8.24h
[04:07:05] Validando...
[04:07:21] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[04:07:39] step=153100 | loss=2.1515 | acc30=0.286 | lr=1.00e-08 | 8.25h
[04:07:57] step=153200 | loss=2.4147 | acc30=0.192 | lr=1.00e-08 | 8.26h
[04:08:15] step=153300 | loss=2.5861 | acc30=0.165 | lr=1.00e-08 | 8.26h
[04:08:33] step=153400 | loss=2.9892 | acc30=0.068 | lr=1.00e-08 | 8.27h
[04:08:50] step=153500 | loss=2.4245 | acc30=0.220 | lr=1.00e-08 | 8.27h
[04:09:08] step=153600 | loss=3.1370 | acc30=0.069 | lr=1.00e-08 | 8.28h
[04:09:26] step=153700 | loss=2.5458 | acc30=0.169 | lr=1.00e-08 | 8.28h
[04:09:44] step=153800 | loss=2.2789 | acc30=0.208 | lr=1.00e-08 | 8.29h
[04:10:01] step=153900 | loss=3.3735 | acc30=0.054 | lr=1.00e-08 | 8.29h
[04:10:19] step=154000 | loss=2.2983 | acc30=0.268 | lr=1.00e-08 | 8.30h
[04:10:19] Validando...
[04:10:35] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:10:53] step=154100 | loss=2.1503 | acc30=0.291 | lr=1.00e-08 | 8.31h
[04:11:11] step=154200 | loss=2.1757 | acc30=0.273 | lr=1.00e-08 | 8.31h
[04:11:28] step=154300 | loss=2.1012 | acc30=0.340 | lr=1.00e-08 | 8.32h
[04:11:46] step=154400 | loss=2.8867 | acc30=0.146 | lr=1.00e-08 | 8.32h
[04:12:04] step=154500 | loss=2.2365 | acc30=0.265 | lr=1.00e-08 | 8.33h
[04:12:22] step=154600 | loss=2.0862 | acc30=0.347 | lr=1.00e-08 | 8.33h
[04:12:39] step=154700 | loss=2.4762 | acc30=0.148 | lr=1.00e-08 | 8.34h
[04:12:57] step=154800 | loss=2.7639 | acc30=0.108 | lr=1.00e-08 | 8.34h
[04:13:15] step=154900 | loss=2.6007 | acc30=0.130 | lr=1.00e-08 | 8.35h
[04:13:32] step=155000 | loss=2.3176 | acc30=0.197 | lr=1.00e-08 | 8.35h
[04:13:32] Validando...
[04:13:49] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[04:14:07] step=155100 | loss=2.2351 | acc30=0.231 | lr=1.00e-08 | 8.36h
[04:14:25] step=155200 | loss=2.5028 | acc30=0.145 | lr=1.00e-08 | 8.37h
[04:14:42] step=155300 | loss=2.4153 | acc30=0.184 | lr=1.00e-08 | 8.37h
[04:15:00] step=155400 | loss=2.2218 | acc30=0.242 | lr=1.00e-08 | 8.38h
[04:15:18] step=155500 | loss=2.0771 | acc30=0.344 | lr=1.00e-08 | 8.38h
[04:15:36] step=155600 | loss=2.1424 | acc30=0.293 | lr=1.00e-08 | 8.39h
[04:15:54] step=155700 | loss=2.1204 | acc30=0.311 | lr=1.00e-08 | 8.39h
[04:16:11] step=155800 | loss=2.1473 | acc30=0.303 | lr=1.00e-08 | 8.40h
[04:16:29] step=155900 | loss=2.7377 | acc30=0.127 | lr=1.00e-08 | 8.40h
[04:16:47] step=156000 | loss=2.2517 | acc30=0.244 | lr=1.00e-08 | 8.41h
[04:16:47] Validando...
[04:17:03] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:17:21] step=156100 | loss=2.2267 | acc30=0.244 | lr=1.00e-08 | 8.42h
[04:17:38] step=156200 | loss=2.6142 | acc30=0.149 | lr=1.00e-08 | 8.42h
[04:17:56] step=156300 | loss=4.3377 | acc30=0.041 | lr=1.00e-08 | 8.43h
[04:18:14] step=156400 | loss=3.4630 | acc30=0.072 | lr=1.00e-08 | 8.43h
[04:18:32] step=156500 | loss=2.1993 | acc30=0.256 | lr=1.00e-08 | 8.44h
[04:18:49] step=156600 | loss=2.2213 | acc30=0.232 | lr=1.00e-08 | 8.44h
[04:19:07] step=156700 | loss=3.6250 | acc30=0.064 | lr=1.00e-08 | 8.45h
[04:19:25] step=156800 | loss=2.2997 | acc30=0.228 | lr=1.00e-08 | 8.45h
[04:19:43] step=156900 | loss=2.1480 | acc30=0.292 | lr=1.00e-08 | 8.46h
[04:20:00] step=157000 | loss=2.2102 | acc30=0.244 | lr=1.00e-08 | 8.46h
[04:20:00] Validando...
[04:20:16] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:20:34] step=157100 | loss=2.2208 | acc30=0.242 | lr=1.00e-08 | 8.47h
[04:20:53] step=157200 | loss=2.2650 | acc30=0.234 | lr=1.00e-08 | 8.47h
[04:21:11] step=157300 | loss=2.3064 | acc30=0.239 | lr=1.00e-08 | 8.48h
[04:21:28] step=157400 | loss=2.8629 | acc30=0.086 | lr=1.00e-08 | 8.48h
[04:21:46] step=157500 | loss=2.1964 | acc30=0.297 | lr=1.00e-08 | 8.49h
[04:22:04] step=157600 | loss=3.0003 | acc30=0.198 | lr=1.00e-08 | 8.49h
[04:22:22] step=157700 | loss=3.1859 | acc30=0.080 | lr=1.00e-08 | 8.50h
[04:22:39] step=157800 | loss=2.4908 | acc30=0.148 | lr=1.00e-08 | 8.50h
[04:22:57] step=157900 | loss=2.2987 | acc30=0.205 | lr=1.00e-08 | 8.51h
[04:23:15] step=158000 | loss=2.2748 | acc30=0.224 | lr=1.00e-08 | 8.51h
[04:23:15] Validando...
[04:23:31] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:23:48] step=158100 | loss=2.1417 | acc30=0.307 | lr=1.00e-08 | 8.52h
[04:24:06] step=158200 | loss=2.1227 | acc30=0.318 | lr=1.00e-08 | 8.53h
[04:24:24] step=158300 | loss=2.9820 | acc30=0.090 | lr=1.00e-08 | 8.53h
[04:24:42] step=158400 | loss=2.8480 | acc30=0.106 | lr=1.00e-08 | 8.54h
[04:25:00] step=158500 | loss=2.6929 | acc30=0.143 | lr=1.00e-08 | 8.54h
[04:25:17] step=158600 | loss=2.1739 | acc30=0.275 | lr=1.00e-08 | 8.55h
[04:25:35] step=158700 | loss=2.2069 | acc30=0.260 | lr=1.00e-08 | 8.55h
[04:25:53] step=158800 | loss=2.2509 | acc30=0.223 | lr=1.00e-08 | 8.56h
[04:26:11] step=158900 | loss=2.2460 | acc30=0.224 | lr=1.00e-08 | 8.56h
[04:26:28] step=159000 | loss=2.1279 | acc30=0.302 | lr=1.00e-08 | 8.57h
[04:26:28] Validando...
[04:26:44] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[04:27:03] step=159100 | loss=2.3128 | acc30=0.219 | lr=1.00e-08 | 8.58h
[04:27:21] step=159200 | loss=2.2987 | acc30=0.227 | lr=1.00e-08 | 8.58h
[04:27:39] step=159300 | loss=2.7957 | acc30=0.228 | lr=1.00e-08 | 8.59h
[04:27:56] step=159400 | loss=4.7653 | acc30=0.023 | lr=1.00e-08 | 8.59h
[04:28:14] step=159500 | loss=2.3416 | acc30=0.238 | lr=1.00e-08 | 8.60h
[04:28:32] step=159600 | loss=2.0920 | acc30=0.356 | lr=1.00e-08 | 8.60h
[04:28:50] step=159700 | loss=2.2707 | acc30=0.338 | lr=1.00e-08 | 8.61h
[04:29:07] step=159800 | loss=3.4297 | acc30=0.069 | lr=1.00e-08 | 8.61h
[04:29:25] step=159900 | loss=2.6799 | acc30=0.119 | lr=1.00e-08 | 8.62h
[04:29:43] step=160000 | loss=2.5286 | acc30=0.147 | lr=1.00e-08 | 8.62h
[04:29:43] Validando...
[04:29:59] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:30:17] step=160100 | loss=2.3287 | acc30=0.203 | lr=1.00e-08 | 8.63h
[04:30:34] step=160200 | loss=2.6377 | acc30=0.115 | lr=1.00e-08 | 8.64h
[04:30:52] step=160300 | loss=2.5227 | acc30=0.136 | lr=1.00e-08 | 8.64h
[04:31:10] step=160400 | loss=2.6807 | acc30=0.167 | lr=1.00e-08 | 8.65h
[04:31:28] step=160500 | loss=2.6884 | acc30=0.120 | lr=1.00e-08 | 8.65h
[04:31:45] step=160600 | loss=3.9605 | acc30=0.026 | lr=1.00e-08 | 8.66h
[04:32:03] step=160700 | loss=2.8566 | acc30=0.095 | lr=1.00e-08 | 8.66h
[04:32:21] step=160800 | loss=2.4951 | acc30=0.192 | lr=1.00e-08 | 8.67h
[04:32:39] step=160900 | loss=2.0905 | acc30=0.327 | lr=1.00e-08 | 8.67h
[04:32:56] step=161000 | loss=2.2126 | acc30=0.259 | lr=1.00e-08 | 8.68h
[04:32:56] Validando...
[04:33:12] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:33:30] step=161100 | loss=2.5673 | acc30=0.144 | lr=1.00e-08 | 8.69h
[04:33:48] step=161200 | loss=2.2565 | acc30=0.227 | lr=1.00e-08 | 8.69h
[04:34:06] step=161300 | loss=2.2561 | acc30=0.219 | lr=1.00e-08 | 8.70h
[04:34:23] step=161400 | loss=2.3719 | acc30=0.191 | lr=1.00e-08 | 8.70h
[04:34:41] step=161500 | loss=2.3111 | acc30=0.197 | lr=1.00e-08 | 8.70h
[04:34:59] step=161600 | loss=2.2459 | acc30=0.269 | lr=1.00e-08 | 8.71h
[04:35:17] step=161700 | loss=2.4180 | acc30=0.177 | lr=1.00e-08 | 8.71h
[04:35:35] step=161800 | loss=2.1341 | acc30=0.286 | lr=1.00e-08 | 8.72h
[04:35:53] step=161900 | loss=2.8346 | acc30=0.112 | lr=1.00e-08 | 8.72h
[04:36:11] step=162000 | loss=2.4645 | acc30=0.158 | lr=1.00e-08 | 8.73h
[04:36:11] Validando...
[04:36:27] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[04:36:44] step=162100 | loss=2.2238 | acc30=0.257 | lr=1.00e-08 | 8.74h
[04:37:02] step=162200 | loss=2.3954 | acc30=0.167 | lr=1.00e-08 | 8.74h
[04:37:20] step=162300 | loss=2.2428 | acc30=0.246 | lr=1.00e-08 | 8.75h
[04:37:38] step=162400 | loss=2.2910 | acc30=0.258 | lr=1.00e-08 | 8.75h
[04:37:55] step=162500 | loss=2.7846 | acc30=0.104 | lr=1.00e-08 | 8.76h
[04:38:13] step=162600 | loss=3.9680 | acc30=0.032 | lr=1.00e-08 | 8.76h
[04:38:31] step=162700 | loss=2.6179 | acc30=0.145 | lr=1.00e-08 | 8.77h
[04:38:49] step=162800 | loss=2.3110 | acc30=0.209 | lr=1.00e-08 | 8.77h
[04:39:06] step=162900 | loss=2.4949 | acc30=0.140 | lr=1.00e-08 | 8.78h
[04:39:24] step=163000 | loss=2.2804 | acc30=0.203 | lr=1.00e-08 | 8.78h
[04:39:24] Validando...
[04:39:40] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[04:39:58] step=163100 | loss=2.1782 | acc30=0.268 | lr=1.00e-08 | 8.79h
[04:40:16] step=163200 | loss=2.3868 | acc30=0.181 | lr=1.00e-08 | 8.80h
[04:40:33] step=163300 | loss=2.3467 | acc30=0.185 | lr=1.00e-08 | 8.80h
[04:40:51] step=163400 | loss=2.1363 | acc30=0.318 | lr=1.00e-08 | 8.81h
[04:41:09] step=163500 | loss=2.2221 | acc30=0.270 | lr=1.00e-08 | 8.81h
[04:41:27] step=163600 | loss=2.2215 | acc30=0.243 | lr=1.00e-08 | 8.82h
[04:41:44] step=163700 | loss=2.2237 | acc30=0.264 | lr=1.00e-08 | 8.82h
[04:42:02] step=163800 | loss=2.5354 | acc30=0.162 | lr=1.00e-08 | 8.83h
[04:42:20] step=163900 | loss=3.0757 | acc30=0.093 | lr=1.00e-08 | 8.83h
[04:42:38] step=164000 | loss=3.5466 | acc30=0.094 | lr=1.00e-08 | 8.84h
[04:42:38] Validando...
[04:42:54] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:43:11] step=164100 | loss=2.0714 | acc30=0.376 | lr=1.00e-08 | 8.85h
[04:43:29] step=164200 | loss=2.3355 | acc30=0.215 | lr=1.00e-08 | 8.85h
[04:43:47] step=164300 | loss=3.1126 | acc30=0.106 | lr=1.00e-08 | 8.86h
[04:44:05] step=164400 | loss=3.4326 | acc30=0.071 | lr=1.00e-08 | 8.86h
[04:44:22] step=164500 | loss=2.1506 | acc30=0.292 | lr=1.00e-08 | 8.87h
[04:44:40] step=164600 | loss=2.3087 | acc30=0.216 | lr=1.00e-08 | 8.87h
[04:44:58] step=164700 | loss=2.2050 | acc30=0.272 | lr=1.00e-08 | 8.88h
[04:45:16] step=164800 | loss=2.1435 | acc30=0.311 | lr=1.00e-08 | 8.88h
[04:45:33] step=164900 | loss=3.6907 | acc30=0.036 | lr=1.00e-08 | 8.89h
[04:45:51] step=165000 | loss=2.5599 | acc30=0.129 | lr=1.00e-08 | 8.89h
[04:45:51] Validando...
[04:46:07] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:46:25] step=165100 | loss=2.2546 | acc30=0.245 | lr=1.00e-08 | 8.90h
[04:46:43] step=165200 | loss=2.1884 | acc30=0.267 | lr=1.00e-08 | 8.91h
[04:47:01] step=165300 | loss=2.1365 | acc30=0.295 | lr=1.00e-08 | 8.91h
[04:47:18] step=165400 | loss=2.2901 | acc30=0.209 | lr=1.00e-08 | 8.92h
[04:47:37] step=165500 | loss=2.3982 | acc30=0.171 | lr=1.00e-08 | 8.92h
[04:47:55] step=165600 | loss=2.4859 | acc30=0.155 | lr=1.00e-08 | 8.93h
[04:48:12] step=165700 | loss=2.1816 | acc30=0.267 | lr=1.00e-08 | 8.93h
[04:48:30] step=165800 | loss=2.7384 | acc30=0.110 | lr=1.00e-08 | 8.94h
[04:48:48] step=165900 | loss=2.4157 | acc30=0.221 | lr=1.00e-08 | 8.94h
[04:49:05] step=166000 | loss=2.1473 | acc30=0.304 | lr=1.00e-08 | 8.94h
[04:49:05] Validando...
[04:49:22] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[04:49:40] step=166100 | loss=3.3410 | acc30=0.069 | lr=1.00e-08 | 8.95h
[04:49:58] step=166200 | loss=3.0862 | acc30=0.065 | lr=1.00e-08 | 8.96h
[04:50:16] step=166300 | loss=2.4234 | acc30=0.202 | lr=1.00e-08 | 8.96h
[04:50:33] step=166400 | loss=2.2050 | acc30=0.265 | lr=1.00e-08 | 8.97h
[04:50:51] step=166500 | loss=2.7713 | acc30=0.110 | lr=1.00e-08 | 8.97h
[04:51:09] step=166600 | loss=2.4753 | acc30=0.197 | lr=1.00e-08 | 8.98h
[04:51:27] step=166700 | loss=2.1724 | acc30=0.273 | lr=1.00e-08 | 8.98h
[04:51:44] step=166800 | loss=2.2714 | acc30=0.215 | lr=1.00e-08 | 8.99h
[04:52:02] step=166900 | loss=2.7690 | acc30=0.104 | lr=1.00e-08 | 8.99h
[04:52:20] step=167000 | loss=2.3481 | acc30=0.266 | lr=1.00e-08 | 9.00h
[04:52:20] Validando...
[04:52:36] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:52:36] --- CHECKPOINT (hora 9.0) ---
[04:52:36] Checkpoint salvo: checkpoint_step00167000.pt (59.0 MB)
[04:52:43] ☁ Upload HuggingFace OK: step=167000
[04:52:55] step=167100 | loss=2.2986 | acc30=0.309 | lr=1.00e-08 | 9.01h
[04:53:12] step=167200 | loss=3.0667 | acc30=0.077 | lr=1.00e-08 | 9.01h
[04:53:30] step=167300 | loss=2.0843 | acc30=0.343 | lr=1.00e-08 | 9.02h
[04:53:48] step=167400 | loss=2.2679 | acc30=0.265 | lr=1.00e-08 | 9.02h
[04:54:06] step=167500 | loss=2.3807 | acc30=0.201 | lr=1.00e-08 | 9.03h
[04:54:23] step=167600 | loss=2.5944 | acc30=0.146 | lr=1.00e-08 | 9.03h
[04:54:41] step=167700 | loss=2.1856 | acc30=0.304 | lr=1.00e-08 | 9.04h
[04:54:59] step=167800 | loss=2.2545 | acc30=0.236 | lr=1.00e-08 | 9.04h
[04:55:17] step=167900 | loss=2.2941 | acc30=0.232 | lr=1.00e-08 | 9.05h
[04:55:35] step=168000 | loss=2.0223 | acc30=0.417 | lr=1.00e-08 | 9.05h
[04:55:35] Validando...
[04:55:51] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:56:09] step=168100 | loss=3.7062 | acc30=0.092 | lr=1.00e-08 | 9.06h
[04:56:27] step=168200 | loss=3.3826 | acc30=0.077 | lr=1.00e-08 | 9.07h
[04:56:45] step=168300 | loss=3.3563 | acc30=0.074 | lr=1.00e-08 | 9.07h
[04:57:02] step=168400 | loss=2.3936 | acc30=0.187 | lr=1.00e-08 | 9.08h
[04:57:20] step=168500 | loss=2.2270 | acc30=0.303 | lr=1.00e-08 | 9.08h
[04:57:38] step=168600 | loss=2.6974 | acc30=0.224 | lr=1.00e-08 | 9.09h
[04:57:56] step=168700 | loss=3.2440 | acc30=0.075 | lr=1.00e-08 | 9.09h
[04:58:13] step=168800 | loss=2.2029 | acc30=0.255 | lr=1.00e-08 | 9.10h
[04:58:31] step=168900 | loss=2.2124 | acc30=0.260 | lr=1.00e-08 | 9.10h
[04:58:49] step=169000 | loss=2.4437 | acc30=0.215 | lr=1.00e-08 | 9.11h
[04:58:49] Validando...
[04:59:05] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[04:59:23] step=169100 | loss=2.7552 | acc30=0.110 | lr=1.00e-08 | 9.12h
[04:59:41] step=169200 | loss=2.1680 | acc30=0.285 | lr=1.00e-08 | 9.12h
[04:59:59] step=169300 | loss=2.2005 | acc30=0.275 | lr=1.00e-08 | 9.13h
[05:00:17] step=169400 | loss=2.2433 | acc30=0.244 | lr=1.00e-08 | 9.13h
[05:00:35] step=169500 | loss=2.2966 | acc30=0.218 | lr=1.00e-08 | 9.14h
[05:00:52] step=169600 | loss=2.1617 | acc30=0.286 | lr=1.00e-08 | 9.14h
[05:01:10] step=169700 | loss=2.2235 | acc30=0.252 | lr=1.00e-08 | 9.15h
[05:01:28] step=169800 | loss=2.3635 | acc30=0.202 | lr=1.00e-08 | 9.15h
[05:01:45] step=169900 | loss=2.2504 | acc30=0.245 | lr=1.00e-08 | 9.16h
[05:02:03] step=170000 | loss=2.3244 | acc30=0.199 | lr=1.00e-08 | 9.16h
[05:02:03] Validando...
[05:02:19] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[05:02:37] step=170100 | loss=2.3489 | acc30=0.199 | lr=1.00e-08 | 9.17h
[05:02:55] step=170200 | loss=2.3618 | acc30=0.200 | lr=1.00e-08 | 9.18h
[05:03:13] step=170300 | loss=2.5310 | acc30=0.146 | lr=1.00e-08 | 9.18h
[05:03:31] step=170400 | loss=2.4301 | acc30=0.172 | lr=1.00e-08 | 9.19h
[05:03:49] step=170500 | loss=2.4608 | acc30=0.164 | lr=1.00e-08 | 9.19h
[05:04:07] step=170600 | loss=2.2783 | acc30=0.238 | lr=1.00e-08 | 9.20h
[05:04:24] step=170700 | loss=2.0680 | acc30=0.373 | lr=1.00e-08 | 9.20h
[05:04:42] step=170800 | loss=2.4079 | acc30=0.192 | lr=1.00e-08 | 9.21h
[05:05:00] step=170900 | loss=2.1826 | acc30=0.282 | lr=1.00e-08 | 9.21h
[05:05:18] step=171000 | loss=2.1943 | acc30=0.286 | lr=1.00e-08 | 9.22h
[05:05:18] Validando...
[05:05:34] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[05:05:52] step=171100 | loss=2.2454 | acc30=0.242 | lr=1.00e-08 | 9.22h
[05:06:10] step=171200 | loss=2.1983 | acc30=0.257 | lr=1.00e-08 | 9.23h
[05:06:28] step=171300 | loss=2.2312 | acc30=0.255 | lr=1.00e-08 | 9.23h
[05:06:46] step=171400 | loss=2.2431 | acc30=0.280 | lr=1.00e-08 | 9.24h
[05:07:03] step=171500 | loss=2.3095 | acc30=0.214 | lr=1.00e-08 | 9.24h
[05:07:21] step=171600 | loss=3.3468 | acc30=0.072 | lr=1.00e-08 | 9.25h
[05:07:39] step=171700 | loss=3.3075 | acc30=0.052 | lr=1.00e-08 | 9.25h
[05:07:57] step=171800 | loss=2.8037 | acc30=0.164 | lr=1.00e-08 | 9.26h
[05:08:14] step=171900 | loss=5.5615 | acc30=0.002 | lr=1.00e-08 | 9.26h
[05:08:32] step=172000 | loss=3.6883 | acc30=0.044 | lr=1.00e-08 | 9.27h
[05:08:32] Validando...
[05:08:48] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:09:06] step=172100 | loss=2.9231 | acc30=0.088 | lr=1.00e-08 | 9.28h
[05:09:24] step=172200 | loss=2.5478 | acc30=0.149 | lr=1.00e-08 | 9.28h
[05:09:41] step=172300 | loss=2.4821 | acc30=0.161 | lr=1.00e-08 | 9.29h
[05:09:59] step=172400 | loss=2.1041 | acc30=0.376 | lr=1.00e-08 | 9.29h
[05:10:17] step=172500 | loss=2.2759 | acc30=0.297 | lr=1.00e-08 | 9.30h
[05:10:35] step=172600 | loss=3.1921 | acc30=0.062 | lr=1.00e-08 | 9.30h
[05:10:53] step=172700 | loss=3.5685 | acc30=0.055 | lr=1.00e-08 | 9.31h
[05:11:11] step=172800 | loss=2.9233 | acc30=0.085 | lr=1.00e-08 | 9.31h
[05:11:29] step=172900 | loss=2.7484 | acc30=0.106 | lr=1.00e-08 | 9.32h
[05:11:47] step=173000 | loss=2.6412 | acc30=0.138 | lr=1.00e-08 | 9.32h
[05:11:47] Validando...
[05:12:03] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:12:20] step=173100 | loss=2.2014 | acc30=0.270 | lr=1.00e-08 | 9.33h
[05:12:38] step=173200 | loss=2.2686 | acc30=0.236 | lr=1.00e-08 | 9.34h
[05:12:56] step=173300 | loss=2.4196 | acc30=0.252 | lr=1.00e-08 | 9.34h
[05:13:14] step=173400 | loss=2.8187 | acc30=0.116 | lr=1.00e-08 | 9.35h
[05:13:31] step=173500 | loss=2.0599 | acc30=0.382 | lr=1.00e-08 | 9.35h
[05:13:49] step=173600 | loss=2.5671 | acc30=0.186 | lr=1.00e-08 | 9.36h
[05:14:07] step=173700 | loss=2.9642 | acc30=0.097 | lr=1.00e-08 | 9.36h
[05:14:24] step=173800 | loss=2.5701 | acc30=0.137 | lr=1.00e-08 | 9.37h
[05:14:42] step=173900 | loss=2.5323 | acc30=0.143 | lr=1.00e-08 | 9.37h
[05:15:00] step=174000 | loss=2.1857 | acc30=0.336 | lr=1.00e-08 | 9.38h
[05:15:00] Validando...
[05:15:16] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:15:34] step=174100 | loss=2.1461 | acc30=0.354 | lr=1.00e-08 | 9.39h
[05:15:52] step=174200 | loss=2.1497 | acc30=0.314 | lr=1.00e-08 | 9.39h
[05:16:09] step=174300 | loss=2.1749 | acc30=0.274 | lr=1.00e-08 | 9.40h
[05:16:27] step=174400 | loss=2.1690 | acc30=0.271 | lr=1.00e-08 | 9.40h
[05:16:46] step=174500 | loss=2.2546 | acc30=0.262 | lr=1.00e-08 | 9.41h
[05:17:03] step=174600 | loss=2.3801 | acc30=0.199 | lr=1.00e-08 | 9.41h
[05:17:21] step=174700 | loss=2.1906 | acc30=0.268 | lr=1.00e-08 | 9.42h
[05:17:39] step=174800 | loss=2.7175 | acc30=0.159 | lr=1.00e-08 | 9.42h
[05:17:57] step=174900 | loss=2.6389 | acc30=0.138 | lr=1.00e-08 | 9.43h
[05:18:14] step=175000 | loss=2.0818 | acc30=0.334 | lr=1.00e-08 | 9.43h
[05:18:14] Validando...
[05:18:30] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:18:48] step=175100 | loss=2.3881 | acc30=0.177 | lr=1.00e-08 | 9.44h
[05:19:06] step=175200 | loss=2.5308 | acc30=0.188 | lr=1.00e-08 | 9.45h
[05:19:25] step=175300 | loss=2.6465 | acc30=0.166 | lr=1.00e-08 | 9.45h
[05:19:42] step=175400 | loss=2.0510 | acc30=0.402 | lr=1.00e-08 | 9.46h
[05:20:00] step=175500 | loss=2.0800 | acc30=0.338 | lr=1.00e-08 | 9.46h
[05:20:18] step=175600 | loss=3.0734 | acc30=0.161 | lr=1.00e-08 | 9.47h
[05:20:36] step=175700 | loss=3.5519 | acc30=0.055 | lr=1.00e-08 | 9.47h
[05:20:53] step=175800 | loss=2.8184 | acc30=0.128 | lr=1.00e-08 | 9.47h
[05:21:11] step=175900 | loss=2.2983 | acc30=0.307 | lr=1.00e-08 | 9.48h
[05:21:29] step=176000 | loss=2.4581 | acc30=0.202 | lr=1.00e-08 | 9.48h
[05:21:29] Validando...
[05:21:45] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[05:22:03] step=176100 | loss=2.6347 | acc30=0.111 | lr=1.00e-08 | 9.49h
[05:22:20] step=176200 | loss=2.6490 | acc30=0.116 | lr=1.00e-08 | 9.50h
[05:22:39] step=176300 | loss=3.1760 | acc30=0.085 | lr=1.00e-08 | 9.50h
[05:22:57] step=176400 | loss=2.3849 | acc30=0.182 | lr=1.00e-08 | 9.51h
[05:23:15] step=176500 | loss=2.2505 | acc30=0.226 | lr=1.00e-08 | 9.51h
[05:23:32] step=176600 | loss=2.2329 | acc30=0.247 | lr=1.00e-08 | 9.52h
[05:23:50] step=176700 | loss=2.1196 | acc30=0.324 | lr=1.00e-08 | 9.52h
[05:24:08] step=176800 | loss=2.1098 | acc30=0.316 | lr=1.00e-08 | 9.53h
[05:24:26] step=176900 | loss=2.1400 | acc30=0.297 | lr=1.00e-08 | 9.53h
[05:24:43] step=177000 | loss=2.1223 | acc30=0.330 | lr=1.00e-08 | 9.54h
[05:24:43] Validando...
[05:24:59] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:25:17] step=177100 | loss=2.0593 | acc30=0.406 | lr=1.00e-08 | 9.55h
[05:25:35] step=177200 | loss=2.3532 | acc30=0.260 | lr=1.00e-08 | 9.55h
[05:25:53] step=177300 | loss=2.6226 | acc30=0.123 | lr=1.00e-08 | 9.56h
[05:26:10] step=177400 | loss=2.4623 | acc30=0.172 | lr=1.00e-08 | 9.56h
[05:26:28] step=177500 | loss=2.3731 | acc30=0.176 | lr=1.00e-08 | 9.57h
[05:26:46] step=177600 | loss=2.2391 | acc30=0.248 | lr=1.00e-08 | 9.57h
[05:27:04] step=177700 | loss=2.3831 | acc30=0.187 | lr=1.00e-08 | 9.58h
[05:27:21] step=177800 | loss=2.3628 | acc30=0.208 | lr=1.00e-08 | 9.58h
[05:27:39] step=177900 | loss=2.1974 | acc30=0.283 | lr=1.00e-08 | 9.59h
[05:27:57] step=178000 | loss=2.1618 | acc30=0.282 | lr=1.00e-08 | 9.59h
[05:27:57] Validando...
[05:28:14] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[05:28:32] step=178100 | loss=2.1853 | acc30=0.264 | lr=1.00e-08 | 9.60h
[05:28:49] step=178200 | loss=2.0752 | acc30=0.352 | lr=1.00e-08 | 9.61h
[05:29:07] step=178300 | loss=2.1392 | acc30=0.276 | lr=1.00e-08 | 9.61h
[05:29:25] step=178400 | loss=2.3811 | acc30=0.199 | lr=1.00e-08 | 9.62h
[05:29:43] step=178500 | loss=2.6901 | acc30=0.152 | lr=1.00e-08 | 9.62h
[05:30:00] step=178600 | loss=2.2063 | acc30=0.261 | lr=1.00e-08 | 9.63h
[05:30:18] step=178700 | loss=2.1570 | acc30=0.272 | lr=1.00e-08 | 9.63h
[05:30:36] step=178800 | loss=2.1284 | acc30=0.314 | lr=1.00e-08 | 9.64h
[05:30:54] step=178900 | loss=2.1723 | acc30=0.266 | lr=1.00e-08 | 9.64h
[05:31:11] step=179000 | loss=2.0600 | acc30=0.381 | lr=1.00e-08 | 9.65h
[05:31:11] Validando...
[05:31:28] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:31:46] step=179100 | loss=3.0653 | acc30=0.174 | lr=1.00e-08 | 9.66h
[05:32:04] step=179200 | loss=2.8261 | acc30=0.092 | lr=1.00e-08 | 9.66h
[05:32:21] step=179300 | loss=2.5174 | acc30=0.176 | lr=1.00e-08 | 9.67h
[05:32:39] step=179400 | loss=2.7383 | acc30=0.099 | lr=1.00e-08 | 9.67h
[05:32:57] step=179500 | loss=2.2346 | acc30=0.230 | lr=1.00e-08 | 9.68h
[05:33:15] step=179600 | loss=2.1782 | acc30=0.273 | lr=1.00e-08 | 9.68h
[05:33:32] step=179700 | loss=2.1987 | acc30=0.269 | lr=1.00e-08 | 9.69h
[05:33:50] step=179800 | loss=2.1130 | acc30=0.319 | lr=1.00e-08 | 9.69h
[05:34:08] step=179900 | loss=3.1388 | acc30=0.152 | lr=1.00e-08 | 9.70h
[05:34:26] step=180000 | loss=3.4406 | acc30=0.054 | lr=1.00e-08 | 9.70h
[05:34:26] Validando...
[05:34:42] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:35:00] step=180100 | loss=4.3710 | acc30=0.016 | lr=1.00e-08 | 9.71h
[05:35:18] step=180200 | loss=2.5475 | acc30=0.184 | lr=1.00e-08 | 9.72h
[05:35:36] step=180300 | loss=2.2395 | acc30=0.263 | lr=1.00e-08 | 9.72h
[05:35:54] step=180400 | loss=2.1607 | acc30=0.305 | lr=1.00e-08 | 9.73h
[05:36:11] step=180500 | loss=2.3331 | acc30=0.201 | lr=1.00e-08 | 9.73h
[05:36:29] step=180600 | loss=2.2811 | acc30=0.219 | lr=1.00e-08 | 9.73h
[05:36:47] step=180700 | loss=2.1641 | acc30=0.280 | lr=1.00e-08 | 9.74h
[05:37:05] step=180800 | loss=2.1425 | acc30=0.290 | lr=1.00e-08 | 9.74h
[05:37:22] step=180900 | loss=2.2575 | acc30=0.236 | lr=1.00e-08 | 9.75h
[05:37:40] step=181000 | loss=2.2430 | acc30=0.237 | lr=1.00e-08 | 9.75h
[05:37:40] Validando...
[05:37:56] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:38:14] step=181100 | loss=2.4654 | acc30=0.167 | lr=1.00e-08 | 9.76h
[05:38:32] step=181200 | loss=2.5525 | acc30=0.149 | lr=1.00e-08 | 9.77h
[05:38:49] step=181300 | loss=3.3312 | acc30=0.049 | lr=1.00e-08 | 9.77h
[05:39:07] step=181400 | loss=5.2016 | acc30=0.010 | lr=1.00e-08 | 9.78h
[05:39:25] step=181500 | loss=3.2871 | acc30=0.106 | lr=1.00e-08 | 9.78h
[05:39:43] step=181600 | loss=2.5098 | acc30=0.146 | lr=1.00e-08 | 9.79h
[05:40:01] step=181700 | loss=2.7727 | acc30=0.097 | lr=1.00e-08 | 9.79h
[05:40:18] step=181800 | loss=2.7745 | acc30=0.124 | lr=1.00e-08 | 9.80h
[05:40:36] step=181900 | loss=2.4696 | acc30=0.180 | lr=1.00e-08 | 9.80h
[05:40:54] step=182000 | loss=2.3501 | acc30=0.198 | lr=1.00e-08 | 9.81h
[05:40:54] Validando...
[05:41:10] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[05:41:28] step=182100 | loss=2.2462 | acc30=0.251 | lr=1.00e-08 | 9.82h
[05:41:46] step=182200 | loss=2.1848 | acc30=0.264 | lr=1.00e-08 | 9.82h
[05:42:03] step=182300 | loss=2.1300 | acc30=0.323 | lr=1.00e-08 | 9.83h
[05:42:21] step=182400 | loss=2.1479 | acc30=0.329 | lr=1.00e-08 | 9.83h
[05:42:39] step=182500 | loss=2.8133 | acc30=0.083 | lr=1.00e-08 | 9.84h
[05:42:57] step=182600 | loss=2.4345 | acc30=0.186 | lr=1.00e-08 | 9.84h
[05:43:15] step=182700 | loss=2.2883 | acc30=0.230 | lr=1.00e-08 | 9.85h
[05:43:33] step=182800 | loss=2.4020 | acc30=0.174 | lr=1.00e-08 | 9.85h
[05:43:50] step=182900 | loss=2.3596 | acc30=0.204 | lr=1.00e-08 | 9.86h
[05:44:08] step=183000 | loss=2.2286 | acc30=0.259 | lr=1.00e-08 | 9.86h
[05:44:08] Validando...
[05:44:24] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[05:44:42] step=183100 | loss=2.5534 | acc30=0.218 | lr=1.00e-08 | 9.87h
[05:45:00] step=183200 | loss=2.9539 | acc30=0.095 | lr=1.00e-08 | 9.88h
[05:45:17] step=183300 | loss=2.0523 | acc30=0.393 | lr=1.00e-08 | 9.88h
[05:45:35] step=183400 | loss=2.5277 | acc30=0.211 | lr=1.00e-08 | 9.89h
[05:45:53] step=183500 | loss=2.7245 | acc30=0.122 | lr=1.00e-08 | 9.89h
[05:46:10] step=183600 | loss=2.6925 | acc30=0.157 | lr=1.00e-08 | 9.90h
[05:46:28] step=183700 | loss=2.4564 | acc30=0.158 | lr=1.00e-08 | 9.90h
[05:46:47] step=183800 | loss=3.5333 | acc30=0.057 | lr=1.00e-08 | 9.91h
[05:47:05] step=183900 | loss=3.1748 | acc30=0.087 | lr=1.00e-08 | 9.91h
[05:47:22] step=184000 | loss=3.2453 | acc30=0.075 | lr=1.00e-08 | 9.92h
[05:47:22] Validando...
[05:47:39] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[05:47:56] step=184100 | loss=2.5548 | acc30=0.148 | lr=1.00e-08 | 9.93h
[05:48:14] step=184200 | loss=2.3250 | acc30=0.215 | lr=1.00e-08 | 9.93h
[05:48:32] step=184300 | loss=2.1968 | acc30=0.251 | lr=1.00e-08 | 9.94h
[05:48:49] step=184400 | loss=2.1827 | acc30=0.265 | lr=1.00e-08 | 9.94h
[05:49:07] step=184500 | loss=2.5164 | acc30=0.142 | lr=1.00e-08 | 9.95h
[05:49:25] step=184600 | loss=2.3980 | acc30=0.173 | lr=1.00e-08 | 9.95h
[05:49:44] step=184700 | loss=2.3592 | acc30=0.209 | lr=1.00e-08 | 9.96h
[05:50:01] step=184800 | loss=2.7473 | acc30=0.110 | lr=1.00e-08 | 9.96h
[05:50:19] step=184900 | loss=2.6434 | acc30=0.125 | lr=1.00e-08 | 9.97h
[05:50:37] step=185000 | loss=2.2954 | acc30=0.203 | lr=1.00e-08 | 9.97h
[05:50:37] Validando...
[05:50:53] VAL: loss=2.3523 acc30=0.200 acc60=0.272 acc120=0.337
[05:51:11] step=185100 | loss=2.3358 | acc30=0.206 | lr=1.00e-08 | 9.98h
[05:51:28] step=185200 | loss=2.4498 | acc30=0.189 | lr=1.00e-08 | 9.98h
[05:51:46] step=185300 | loss=2.1591 | acc30=0.276 | lr=1.00e-08 | 9.99h
[05:52:04] step=185400 | loss=2.5200 | acc30=0.191 | lr=1.00e-08 | 9.99h
[05:52:22] step=185500 | loss=2.7507 | acc30=0.108 | lr=1.00e-08 | 10.00h
[05:52:37] --- CHECKPOINT (hora 10.0) ---
[05:52:37] Checkpoint salvo: checkpoint_step00185584.pt (59.0 MB)
[05:52:40] step=185600 | loss=3.1126 | acc30=0.069 | lr=1.00e-08 | 10.00h
[05:52:53] ☁ Upload HuggingFace OK: step=185584
[05:52:58] step=185700 | loss=2.3349 | acc30=0.249 | lr=1.00e-08 | 10.01h
[05:53:16] step=185800 | loss=2.1920 | acc30=0.288 | lr=1.00e-08 | 10.01h
[05:53:34] step=185900 | loss=2.4809 | acc30=0.157 | lr=1.00e-08 | 10.02h
[05:53:51] step=186000 | loss=2.7568 | acc30=0.098 | lr=1.00e-08 | 10.02h
[05:53:51] Validando...
[05:54:07] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[05:54:25] step=186100 | loss=2.5982 | acc30=0.155 | lr=1.00e-08 | 10.03h
[05:54:44] step=186200 | loss=2.2477 | acc30=0.243 | lr=1.00e-08 | 10.04h
[05:55:02] step=186300 | loss=2.4496 | acc30=0.147 | lr=1.00e-08 | 10.04h
[05:55:19] step=186400 | loss=2.8453 | acc30=0.088 | lr=1.00e-08 | 10.05h
[05:55:37] step=186500 | loss=3.4334 | acc30=0.046 | lr=1.00e-08 | 10.05h
[05:55:55] step=186600 | loss=2.9759 | acc30=0.149 | lr=1.00e-08 | 10.06h
[05:56:13] step=186700 | loss=2.1434 | acc30=0.294 | lr=1.00e-08 | 10.06h
[05:56:30] step=186800 | loss=2.3158 | acc30=0.199 | lr=1.00e-08 | 10.07h
[05:56:48] step=186900 | loss=2.4412 | acc30=0.154 | lr=1.00e-08 | 10.07h
[05:57:06] step=187000 | loss=2.3334 | acc30=0.221 | lr=1.00e-08 | 10.08h
[05:57:06] Validando...
[05:57:22] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[05:57:40] step=187100 | loss=2.1016 | acc30=0.323 | lr=1.00e-08 | 10.09h
[05:57:58] step=187200 | loss=2.3327 | acc30=0.194 | lr=1.00e-08 | 10.09h
[05:58:16] step=187300 | loss=2.1622 | acc30=0.310 | lr=1.00e-08 | 10.10h
[05:58:34] step=187400 | loss=2.2813 | acc30=0.300 | lr=1.00e-08 | 10.10h
[05:58:52] step=187500 | loss=2.4332 | acc30=0.179 | lr=1.00e-08 | 10.11h
[05:59:09] step=187600 | loss=2.1282 | acc30=0.311 | lr=1.00e-08 | 10.11h
[05:59:27] step=187700 | loss=2.2129 | acc30=0.256 | lr=1.00e-08 | 10.12h
[05:59:45] step=187800 | loss=2.2491 | acc30=0.231 | lr=1.00e-08 | 10.12h
[06:00:02] step=187900 | loss=2.2436 | acc30=0.227 | lr=1.00e-08 | 10.13h
[06:00:20] step=188000 | loss=2.3144 | acc30=0.207 | lr=1.00e-08 | 10.13h
[06:00:20] Validando...
[06:00:36] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:00:54] step=188100 | loss=2.4783 | acc30=0.148 | lr=1.00e-08 | 10.14h
[06:01:12] step=188200 | loss=2.2654 | acc30=0.247 | lr=1.00e-08 | 10.15h
[06:01:30] step=188300 | loss=2.4102 | acc30=0.198 | lr=1.00e-08 | 10.15h
[06:01:47] step=188400 | loss=4.4526 | acc30=0.019 | lr=1.00e-08 | 10.16h
[06:02:05] step=188500 | loss=2.9793 | acc30=0.166 | lr=1.00e-08 | 10.16h
[06:02:23] step=188600 | loss=2.3788 | acc30=0.182 | lr=1.00e-08 | 10.17h
[06:02:41] step=188700 | loss=2.4327 | acc30=0.154 | lr=1.00e-08 | 10.17h
[06:02:59] step=188800 | loss=2.4433 | acc30=0.154 | lr=1.00e-08 | 10.18h
[06:03:17] step=188900 | loss=3.9525 | acc30=0.072 | lr=1.00e-08 | 10.18h
[06:03:35] step=189000 | loss=3.3693 | acc30=0.106 | lr=1.00e-08 | 10.19h
[06:03:35] Validando...
[06:03:51] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:04:08] step=189100 | loss=2.1131 | acc30=0.319 | lr=1.00e-08 | 10.20h
[06:04:26] step=189200 | loss=2.3220 | acc30=0.188 | lr=1.00e-08 | 10.20h
[06:04:44] step=189300 | loss=2.2274 | acc30=0.256 | lr=1.00e-08 | 10.21h
[06:05:02] step=189400 | loss=2.1079 | acc30=0.326 | lr=1.00e-08 | 10.21h
[06:05:19] step=189500 | loss=2.0857 | acc30=0.365 | lr=1.00e-08 | 10.22h
[06:05:37] step=189600 | loss=2.3795 | acc30=0.182 | lr=1.00e-08 | 10.22h
[06:05:55] step=189700 | loss=2.8678 | acc30=0.125 | lr=1.00e-08 | 10.23h
[06:06:13] step=189800 | loss=2.9315 | acc30=0.131 | lr=1.00e-08 | 10.23h
[06:06:31] step=189900 | loss=2.1986 | acc30=0.269 | lr=1.00e-08 | 10.24h
[06:06:48] step=190000 | loss=3.6419 | acc30=0.061 | lr=1.00e-08 | 10.24h
[06:06:48] Validando...
[06:07:05] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:07:23] step=190100 | loss=2.7211 | acc30=0.126 | lr=1.00e-08 | 10.25h
[06:07:41] step=190200 | loss=2.2154 | acc30=0.264 | lr=1.00e-08 | 10.25h
[06:07:58] step=190300 | loss=2.1933 | acc30=0.268 | lr=1.00e-08 | 10.26h
[06:08:16] step=190400 | loss=2.3512 | acc30=0.183 | lr=1.00e-08 | 10.26h
[06:08:34] step=190500 | loss=2.2779 | acc30=0.211 | lr=1.00e-08 | 10.27h
[06:08:52] step=190600 | loss=2.2286 | acc30=0.251 | lr=1.00e-08 | 10.27h
[06:09:09] step=190700 | loss=2.1885 | acc30=0.264 | lr=1.00e-08 | 10.28h
[06:09:27] step=190800 | loss=2.4472 | acc30=0.166 | lr=1.00e-08 | 10.28h
[06:09:45] step=190900 | loss=2.4226 | acc30=0.172 | lr=1.00e-08 | 10.29h
[06:10:03] step=191000 | loss=2.2353 | acc30=0.253 | lr=1.00e-08 | 10.29h
[06:10:03] Validando...
[06:10:19] VAL: loss=2.3523 acc30=0.200 acc60=0.272 acc120=0.337
[06:10:36] step=191100 | loss=2.2153 | acc30=0.256 | lr=1.00e-08 | 10.30h
[06:10:54] step=191200 | loss=2.7077 | acc30=0.113 | lr=1.00e-08 | 10.31h
[06:11:12] step=191300 | loss=2.8185 | acc30=0.186 | lr=1.00e-08 | 10.31h
[06:11:30] step=191400 | loss=2.2702 | acc30=0.277 | lr=1.00e-08 | 10.32h
[06:11:47] step=191500 | loss=2.7711 | acc30=0.110 | lr=1.00e-08 | 10.32h
[06:12:05] step=191600 | loss=2.8194 | acc30=0.094 | lr=1.00e-08 | 10.33h
[06:12:23] step=191700 | loss=2.6572 | acc30=0.184 | lr=1.00e-08 | 10.33h
[06:12:41] step=191800 | loss=2.1426 | acc30=0.292 | lr=1.00e-08 | 10.34h
[06:12:59] step=191900 | loss=2.4472 | acc30=0.187 | lr=1.00e-08 | 10.34h
[06:13:17] step=192000 | loss=2.1363 | acc30=0.296 | lr=1.00e-08 | 10.35h
[06:13:17] Validando...
[06:13:33] VAL: loss=2.3523 acc30=0.200 acc60=0.272 acc120=0.337
[06:13:51] step=192100 | loss=2.8335 | acc30=0.188 | lr=1.00e-08 | 10.36h
[06:14:09] step=192200 | loss=2.8343 | acc30=0.130 | lr=1.00e-08 | 10.36h
[06:14:26] step=192300 | loss=2.6860 | acc30=0.119 | lr=1.00e-08 | 10.37h
[06:14:44] step=192400 | loss=2.3706 | acc30=0.214 | lr=1.00e-08 | 10.37h
[06:15:02] step=192500 | loss=2.5286 | acc30=0.198 | lr=1.00e-08 | 10.38h
[06:15:20] step=192600 | loss=2.4689 | acc30=0.164 | lr=1.00e-08 | 10.38h
[06:15:38] step=192700 | loss=2.0959 | acc30=0.347 | lr=1.00e-08 | 10.39h
[06:15:56] step=192800 | loss=2.3962 | acc30=0.187 | lr=1.00e-08 | 10.39h
[06:16:13] step=192900 | loss=2.1631 | acc30=0.287 | lr=1.00e-08 | 10.40h
[06:16:32] step=193000 | loss=2.0453 | acc30=0.382 | lr=1.00e-08 | 10.40h
[06:16:32] Validando...
[06:16:48] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:17:05] step=193100 | loss=2.2764 | acc30=0.240 | lr=1.00e-08 | 10.41h
[06:17:23] step=193200 | loss=2.1928 | acc30=0.267 | lr=1.00e-08 | 10.42h
[06:17:41] step=193300 | loss=2.2517 | acc30=0.241 | lr=1.00e-08 | 10.42h
[06:17:58] step=193400 | loss=2.4645 | acc30=0.150 | lr=1.00e-08 | 10.43h
[06:18:16] step=193500 | loss=2.2868 | acc30=0.235 | lr=1.00e-08 | 10.43h
[06:18:34] step=193600 | loss=2.5683 | acc30=0.134 | lr=1.00e-08 | 10.44h
[06:18:52] step=193700 | loss=2.4382 | acc30=0.172 | lr=1.00e-08 | 10.44h
[06:19:10] step=193800 | loss=2.2428 | acc30=0.247 | lr=1.00e-08 | 10.45h
[06:19:28] step=193900 | loss=2.1820 | acc30=0.278 | lr=1.00e-08 | 10.45h
[06:19:46] step=194000 | loss=2.6071 | acc30=0.151 | lr=1.00e-08 | 10.46h
[06:19:46] Validando...
[06:20:02] VAL: loss=2.3523 acc30=0.200 acc60=0.272 acc120=0.337
[06:20:19] step=194100 | loss=2.6295 | acc30=0.130 | lr=1.00e-08 | 10.47h
[06:20:37] step=194200 | loss=2.2457 | acc30=0.242 | lr=1.00e-08 | 10.47h
[06:20:55] step=194300 | loss=2.3031 | acc30=0.216 | lr=1.00e-08 | 10.48h
[06:21:12] step=194400 | loss=2.1490 | acc30=0.297 | lr=1.00e-08 | 10.48h
[06:21:30] step=194500 | loss=2.2266 | acc30=0.260 | lr=1.00e-08 | 10.49h
[06:21:48] step=194600 | loss=2.5061 | acc30=0.171 | lr=1.00e-08 | 10.49h
[06:22:06] step=194700 | loss=3.0601 | acc30=0.073 | lr=1.00e-08 | 10.49h
[06:22:24] step=194800 | loss=2.4242 | acc30=0.177 | lr=1.00e-08 | 10.50h
[06:22:41] step=194900 | loss=2.5203 | acc30=0.156 | lr=1.00e-08 | 10.50h
[06:23:00] step=195000 | loss=2.6113 | acc30=0.147 | lr=1.00e-08 | 10.51h
[06:23:00] Validando...
[06:23:16] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:23:33] step=195100 | loss=3.4889 | acc30=0.040 | lr=1.00e-08 | 10.52h
[06:23:51] step=195200 | loss=3.5757 | acc30=0.046 | lr=1.00e-08 | 10.52h
[06:24:09] step=195300 | loss=2.9117 | acc30=0.131 | lr=1.00e-08 | 10.53h
[06:24:27] step=195400 | loss=2.1897 | acc30=0.250 | lr=1.00e-08 | 10.53h
[06:24:44] step=195500 | loss=2.3075 | acc30=0.215 | lr=1.00e-08 | 10.54h
[06:25:02] step=195600 | loss=2.1821 | acc30=0.270 | lr=1.00e-08 | 10.54h
[06:25:20] step=195700 | loss=2.1400 | acc30=0.291 | lr=1.00e-08 | 10.55h
[06:25:38] step=195800 | loss=2.1539 | acc30=0.285 | lr=1.00e-08 | 10.55h
[06:25:55] step=195900 | loss=2.1943 | acc30=0.276 | lr=1.00e-08 | 10.56h
[06:26:13] step=196000 | loss=2.2076 | acc30=0.247 | lr=1.00e-08 | 10.56h
[06:26:13] Validando...
[06:26:29] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:26:47] step=196100 | loss=2.1780 | acc30=0.272 | lr=1.00e-08 | 10.57h
[06:27:04] === Época 2 | step=196196 | elapsed=10.58h ===
[06:27:06] step=196200 | loss=2.1837 | acc30=0.278 | lr=1.00e-08 | 10.58h
[06:27:23] step=196300 | loss=2.0995 | acc30=0.340 | lr=1.00e-08 | 10.58h
[06:27:41] step=196400 | loss=2.3160 | acc30=0.207 | lr=1.00e-08 | 10.59h
[06:27:59] step=196500 | loss=2.5408 | acc30=0.135 | lr=1.00e-08 | 10.59h
[06:28:17] step=196600 | loss=2.2321 | acc30=0.216 | lr=1.00e-08 | 10.60h
[06:28:35] step=196700 | loss=2.9247 | acc30=0.118 | lr=1.00e-08 | 10.60h
[06:28:53] step=196800 | loss=3.9418 | acc30=0.042 | lr=1.00e-08 | 10.61h
[06:29:11] step=196900 | loss=2.5649 | acc30=0.141 | lr=1.00e-08 | 10.61h
[06:29:28] step=197000 | loss=2.8450 | acc30=0.097 | lr=1.00e-08 | 10.62h
[06:29:28] Validando...
[06:29:45] VAL: loss=2.3523 acc30=0.200 acc60=0.272 acc120=0.337
[06:30:02] step=197100 | loss=2.1062 | acc30=0.314 | lr=1.00e-08 | 10.63h
[06:30:20] step=197200 | loss=2.1929 | acc30=0.266 | lr=1.00e-08 | 10.63h
[06:30:38] step=197300 | loss=2.2769 | acc30=0.267 | lr=1.00e-08 | 10.64h
[06:30:55] step=197400 | loss=2.4036 | acc30=0.178 | lr=1.00e-08 | 10.64h
[06:31:13] step=197500 | loss=2.1094 | acc30=0.330 | lr=1.00e-08 | 10.65h
[06:31:31] step=197600 | loss=2.6211 | acc30=0.189 | lr=1.00e-08 | 10.65h
[06:31:49] step=197700 | loss=2.4942 | acc30=0.160 | lr=1.00e-08 | 10.66h
[06:32:07] step=197800 | loss=2.5357 | acc30=0.135 | lr=1.00e-08 | 10.66h
[06:32:25] step=197900 | loss=2.2610 | acc30=0.220 | lr=1.00e-08 | 10.67h
[06:32:43] step=198000 | loss=2.3011 | acc30=0.207 | lr=1.00e-08 | 10.67h
[06:32:43] Validando...
[06:32:59] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:33:17] step=198100 | loss=2.3258 | acc30=0.202 | lr=1.00e-08 | 10.68h
[06:33:34] step=198200 | loss=2.4227 | acc30=0.167 | lr=1.00e-08 | 10.69h
[06:33:52] step=198300 | loss=2.2063 | acc30=0.262 | lr=1.00e-08 | 10.69h
[06:34:10] step=198400 | loss=2.1793 | acc30=0.272 | lr=1.00e-08 | 10.70h
[06:34:28] step=198500 | loss=2.1204 | acc30=0.317 | lr=1.00e-08 | 10.70h
[06:34:46] step=198600 | loss=2.2226 | acc30=0.248 | lr=1.00e-08 | 10.71h
[06:35:03] step=198700 | loss=2.1569 | acc30=0.284 | lr=1.00e-08 | 10.71h
[06:35:21] step=198800 | loss=2.4495 | acc30=0.171 | lr=1.00e-08 | 10.72h
[06:35:39] step=198900 | loss=3.3165 | acc30=0.114 | lr=1.00e-08 | 10.72h
[06:35:57] step=199000 | loss=4.2592 | acc30=0.027 | lr=1.00e-08 | 10.73h
[06:35:57] Validando...
[06:36:13] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:36:31] step=199100 | loss=2.4648 | acc30=0.148 | lr=1.00e-08 | 10.74h
[06:36:49] step=199200 | loss=2.5953 | acc30=0.127 | lr=1.00e-08 | 10.74h
[06:37:07] step=199300 | loss=2.1515 | acc30=0.303 | lr=1.00e-08 | 10.75h
[06:37:25] step=199400 | loss=2.0788 | acc30=0.364 | lr=1.00e-08 | 10.75h
[06:37:42] step=199500 | loss=2.8694 | acc30=0.120 | lr=1.00e-08 | 10.76h
[06:38:00] step=199600 | loss=2.4453 | acc30=0.149 | lr=1.00e-08 | 10.76h
[06:38:18] step=199700 | loss=2.3762 | acc30=0.211 | lr=1.00e-08 | 10.77h
[06:38:36] step=199800 | loss=2.2657 | acc30=0.239 | lr=1.00e-08 | 10.77h
[06:38:53] step=199900 | loss=2.1296 | acc30=0.327 | lr=1.00e-08 | 10.77h
[06:39:11] step=200000 | loss=2.9112 | acc30=0.138 | lr=1.00e-08 | 10.78h
[06:39:11] Validando...
[06:39:28] VAL: loss=2.3523 acc30=0.200 acc60=0.271 acc120=0.337
[06:39:46] step=200100 | loss=2.4830 | acc30=0.184 | lr=1.00e-08 | 10.79h
[06:40:03] step=200200 | loss=2.1881 | acc30=0.255 | lr=1.00e-08 | 10.79h
[06:40:21] step=200300 | loss=2.1498 | acc30=0.272 | lr=1.00e-08 | 10.80h
[06:40:39] step=200400 | loss=2.2513 | acc30=0.214 | lr=1.00e-08 | 10.80h
[06:40:57] step=200500 | loss=2.4471 | acc30=0.184 | lr=1.00e-08 | 10.81h
[06:41:14] step=200600 | loss=2.6582 | acc30=0.117 | lr=1.00e-08 | 10.81h
[06:41:32] step=200700 | loss=2.0720 | acc30=0.368 | lr=1.00e-08 | 10.82h
[06:41:50] step=200800 | loss=2.1112 | acc30=0.331 | lr=1.00e-08 | 10.82h
[06:42:08] step=200900 | loss=2.2519 | acc30=0.251 | lr=1.00e-08 | 10.83h
[06:42:25] step=201000 | loss=2.4253 | acc30=0.177 | lr=1.00e-08 | 10.83h
[06:42:25] Validando...
[06:42:41] VAL: loss=2.3523 acc30=0.200 acc60=0.272 acc120=0.337
[06:42:59] step=201100 | loss=2.3230 | acc30=0.225 | lr=1.00e-08 | 10.84h
[06:43:17] step=201200 | loss=2.1900 | acc30=0.264 | lr=1.00e-08 | 10.85h
[06:43:35] step=201300 | loss=2.0997 | acc30=0.343 | lr=1.00e-08 | 10.85h
[06:43:53] step=201400 | loss=2.1775 | acc30=0.276 | lr=1.00e-08 | 10.86h
[06:44:11] step=201500 | loss=2.1981 | acc30=0.251 | lr=1.00e-08 | 10.86h
[06:44:28] step=201600 | loss=2.4094 | acc30=0.177 | lr=1.00e-08 | 10.87h
[06:44:46] step=201700 | loss=2.4394 | acc30=0.175 | lr=1.00e-08 | 10.87h
[06:45:04] step=201800 | loss=2.3554 | acc30=0.196 | lr=1.00e-08 | 10.88h
[06:45:21] step=201900 | loss=2.1138 | acc30=0.322 | lr=1.00e-08 | 10.88h
[06:45:39] step=202000 | loss=2.1676 | acc30=0.272 | lr=1.00e-08 | 10.89h
[06:45:39] Validando...
[06:45:55] VAL: loss=2.3524 acc30=0.200 acc60=0.272 acc120=0.337
[06:46:13] step=202100 | loss=2.1586 | acc30=0.294 | lr=1.00e-08 | 10.90h
[06:46:32] step=202200 | loss=2.1494 | acc30=0.308 | lr=1.00e-08 | 10.90h
[06:46:49] step=202300 | loss=3.0115 | acc30=0.089 | lr=1.00e-08 | 10.91h
[06:47:07] step=202400 | loss=2.6549 | acc30=0.119 | lr=1.00e-08 | 10.91h
[06:47:24] step=202500 | loss=2.2022 | acc30=0.288 | lr=1.00e-08 | 10.92h
[06:47:42] step=202600 | loss=2.0662 | acc30=0.367 | lr=1.00e-08 | 10.92h
[06:48:00] step=202700 | loss=2.1566 | acc30=0.271 | lr=1.00e-08 | 10.93h
[06:48:18] step=202800 | loss=2.2190 | acc30=0.238 | lr=1.00e-08 | 10.93h
[06:48:36] step=202900 | loss=2.1470 | acc30=0.289 | lr=1.00e-08 | 10.94h
[06:48:54] step=203000 | loss=2.1249 | acc30=0.324 | lr=1.00e-08 | 10.94h
[06:48:54] Validando...
[06:49:10] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[06:49:27] step=203100 | loss=2.4208 | acc30=0.172 | lr=1.00e-08 | 10.95h
[06:49:45] step=203200 | loss=2.3542 | acc30=0.194 | lr=1.00e-08 | 10.96h
[06:50:03] step=203300 | loss=2.2102 | acc30=0.253 | lr=1.00e-08 | 10.96h
[06:50:21] step=203400 | loss=2.2530 | acc30=0.221 | lr=1.00e-08 | 10.97h
[06:50:39] step=203500 | loss=2.3766 | acc30=0.189 | lr=1.00e-08 | 10.97h
[06:50:57] step=203600 | loss=4.3095 | acc30=0.055 | lr=1.00e-08 | 10.98h
[06:51:15] step=203700 | loss=4.5841 | acc30=0.049 | lr=1.00e-08 | 10.98h
[06:51:32] step=203800 | loss=2.3498 | acc30=0.187 | lr=1.00e-08 | 10.99h
[06:51:50] step=203900 | loss=2.6259 | acc30=0.139 | lr=1.00e-08 | 10.99h
[06:52:08] step=204000 | loss=2.3165 | acc30=0.237 | lr=1.00e-08 | 11.00h
[06:52:08] Validando...
[06:52:24] VAL: loss=2.3524 acc30=0.200 acc60=0.271 acc120=0.337
[06:52:24] Tempo máximo atingido (11h). Encerrando.
[06:52:24] === SALVANDO CHECKPOINT FINAL ===
[06:52:24] Validação final...
[06:53:39] VAL FINAL: {'val_loss': 2.443585783958435, 'acc30': np.float64(0.17792578125), 'acc60': np.float64(0.24640234375), 'acc120': np.float64(0.31719140625)}
[06:53:39] Checkpoint salvo: checkpoint_step00204000.pt (59.0 MB)
[06:53:40] Aguardando uploads pendentes...