upload trained GPN MSA model
Browse files- README.md +261 -0
- all_results.json +13 -0
- checkpoint-8800/config.json +27 -0
- checkpoint-8800/optimizer.pt +3 -0
- checkpoint-8800/pytorch_model.bin +3 -0
- checkpoint-8800/rng_state.pth +3 -0
- checkpoint-8800/scheduler.pt +3 -0
- checkpoint-8800/trainer_state.json +2661 -0
- checkpoint-8800/training_args.bin +3 -0
- config.json +27 -0
- eval_results.json +8 -0
- pytorch_model.bin +3 -0
- train_results.json +8 -0
- trainer_state.json +3030 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,261 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
datasets:
|
| 5 |
+
- songlab/gpn-msa-sapiens-dataset
|
| 6 |
+
model-index:
|
| 7 |
+
- name: checkpoints
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# checkpoints
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on the songlab/gpn-msa-sapiens-dataset dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.1593
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 0.0001
|
| 38 |
+
- train_batch_size: 1024
|
| 39 |
+
- eval_batch_size: 1024
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- gradient_accumulation_steps: 2
|
| 43 |
+
- total_train_batch_size: 2048
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: cosine
|
| 46 |
+
- lr_scheduler_warmup_steps: 100
|
| 47 |
+
- training_steps: 10000
|
| 48 |
+
- mixed_precision_training: Native AMP
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 54 |
+
| 0.6225 | 0.0232 | 50 | 0.1988 |
|
| 55 |
+
| 0.1508 | 0.0464 | 100 | 0.1836 |
|
| 56 |
+
| 0.1449 | 0.0696 | 150 | 0.1789 |
|
| 57 |
+
| 0.1421 | 0.0927 | 200 | 0.1762 |
|
| 58 |
+
| 0.141 | 0.1159 | 250 | 0.1764 |
|
| 59 |
+
| 0.1397 | 0.1391 | 300 | 0.1755 |
|
| 60 |
+
| 0.1393 | 0.1623 | 350 | 0.1741 |
|
| 61 |
+
| 0.1388 | 0.1855 | 400 | 0.1738 |
|
| 62 |
+
| 0.1394 | 0.2087 | 450 | 0.1725 |
|
| 63 |
+
| 0.1383 | 0.2319 | 500 | 0.1730 |
|
| 64 |
+
| 0.1376 | 0.2550 | 550 | 0.1717 |
|
| 65 |
+
| 0.1372 | 0.2782 | 600 | 0.1708 |
|
| 66 |
+
| 0.1361 | 0.3014 | 650 | 0.1728 |
|
| 67 |
+
| 0.1362 | 0.3246 | 700 | 0.1715 |
|
| 68 |
+
| 0.1364 | 0.3478 | 750 | 0.1713 |
|
| 69 |
+
| 0.1356 | 0.3710 | 800 | 0.1695 |
|
| 70 |
+
| 0.1353 | 0.3942 | 850 | 0.1687 |
|
| 71 |
+
| 0.1361 | 0.4173 | 900 | 0.1703 |
|
| 72 |
+
| 0.1354 | 0.4405 | 950 | 0.1697 |
|
| 73 |
+
| 0.1352 | 0.4637 | 1000 | 0.1695 |
|
| 74 |
+
| 0.1335 | 0.4869 | 1050 | 0.1683 |
|
| 75 |
+
| 0.1327 | 0.5101 | 1100 | 0.1686 |
|
| 76 |
+
| 0.1337 | 0.5333 | 1150 | 0.1692 |
|
| 77 |
+
| 0.134 | 0.5565 | 1200 | 0.1665 |
|
| 78 |
+
| 0.1341 | 0.5796 | 1250 | 0.1680 |
|
| 79 |
+
| 0.1347 | 0.6028 | 1300 | 0.1672 |
|
| 80 |
+
| 0.1335 | 0.6260 | 1350 | 0.1661 |
|
| 81 |
+
| 0.1338 | 0.6492 | 1400 | 0.1663 |
|
| 82 |
+
| 0.1335 | 0.6724 | 1450 | 0.1670 |
|
| 83 |
+
| 0.1332 | 0.6956 | 1500 | 0.1652 |
|
| 84 |
+
| 0.1336 | 0.7188 | 1550 | 0.1663 |
|
| 85 |
+
| 0.133 | 0.7419 | 1600 | 0.1656 |
|
| 86 |
+
| 0.1332 | 0.7651 | 1650 | 0.1661 |
|
| 87 |
+
| 0.1327 | 0.7883 | 1700 | 0.1656 |
|
| 88 |
+
| 0.1318 | 0.8115 | 1750 | 0.1662 |
|
| 89 |
+
| 0.1319 | 0.8347 | 1800 | 0.1652 |
|
| 90 |
+
| 0.1337 | 0.8579 | 1850 | 0.1639 |
|
| 91 |
+
| 0.1324 | 0.8811 | 1900 | 0.1648 |
|
| 92 |
+
| 0.1334 | 0.9042 | 1950 | 0.1651 |
|
| 93 |
+
| 0.1317 | 0.9274 | 2000 | 0.1638 |
|
| 94 |
+
| 0.1324 | 0.9506 | 2050 | 0.1649 |
|
| 95 |
+
| 0.1326 | 0.9738 | 2100 | 0.1660 |
|
| 96 |
+
| 0.1326 | 0.9970 | 2150 | 0.1640 |
|
| 97 |
+
| 0.132 | 1.0202 | 2200 | 0.1653 |
|
| 98 |
+
| 0.1319 | 1.0434 | 2250 | 0.1655 |
|
| 99 |
+
| 0.1326 | 1.0665 | 2300 | 0.1643 |
|
| 100 |
+
| 0.1321 | 1.0897 | 2350 | 0.1659 |
|
| 101 |
+
| 0.1317 | 1.1129 | 2400 | 0.1644 |
|
| 102 |
+
| 0.1322 | 1.1361 | 2450 | 0.1651 |
|
| 103 |
+
| 0.1325 | 1.1593 | 2500 | 0.1640 |
|
| 104 |
+
| 0.1311 | 1.1825 | 2550 | 0.1626 |
|
| 105 |
+
| 0.1323 | 1.2057 | 2600 | 0.1626 |
|
| 106 |
+
| 0.1316 | 1.2288 | 2650 | 0.1639 |
|
| 107 |
+
| 0.1314 | 1.2520 | 2700 | 0.1635 |
|
| 108 |
+
| 0.1314 | 1.2752 | 2750 | 0.1636 |
|
| 109 |
+
| 0.131 | 1.2984 | 2800 | 0.1626 |
|
| 110 |
+
| 0.1313 | 1.3216 | 2850 | 0.1632 |
|
| 111 |
+
| 0.1312 | 1.3448 | 2900 | 0.1637 |
|
| 112 |
+
| 0.1317 | 1.3680 | 2950 | 0.1640 |
|
| 113 |
+
| 0.1311 | 1.3911 | 3000 | 0.1621 |
|
| 114 |
+
| 0.1304 | 1.4143 | 3050 | 0.1631 |
|
| 115 |
+
| 0.1307 | 1.4375 | 3100 | 0.1624 |
|
| 116 |
+
| 0.1315 | 1.4607 | 3150 | 0.1642 |
|
| 117 |
+
| 0.1303 | 1.4839 | 3200 | 0.1636 |
|
| 118 |
+
| 0.1315 | 1.5071 | 3250 | 0.1622 |
|
| 119 |
+
| 0.1315 | 1.5303 | 3300 | 0.1629 |
|
| 120 |
+
| 0.1303 | 1.5534 | 3350 | 0.1642 |
|
| 121 |
+
| 0.1309 | 1.5766 | 3400 | 0.1618 |
|
| 122 |
+
| 0.1307 | 1.5998 | 3450 | 0.1631 |
|
| 123 |
+
| 0.1314 | 1.6230 | 3500 | 0.1629 |
|
| 124 |
+
| 0.1314 | 1.6462 | 3550 | 0.1628 |
|
| 125 |
+
| 0.1312 | 1.6694 | 3600 | 0.1631 |
|
| 126 |
+
| 0.1299 | 1.6926 | 3650 | 0.1618 |
|
| 127 |
+
| 0.1304 | 1.7157 | 3700 | 0.1624 |
|
| 128 |
+
| 0.1299 | 1.7389 | 3750 | 0.1632 |
|
| 129 |
+
| 0.1309 | 1.7621 | 3800 | 0.1623 |
|
| 130 |
+
| 0.1303 | 1.7853 | 3850 | 0.1631 |
|
| 131 |
+
| 0.1312 | 1.8085 | 3900 | 0.1616 |
|
| 132 |
+
| 0.1303 | 1.8317 | 3950 | 0.1622 |
|
| 133 |
+
| 0.1308 | 1.8549 | 4000 | 0.1632 |
|
| 134 |
+
| 0.1297 | 1.8780 | 4050 | 0.1620 |
|
| 135 |
+
| 0.1301 | 1.9012 | 4100 | 0.1617 |
|
| 136 |
+
| 0.131 | 1.9244 | 4150 | 0.1597 |
|
| 137 |
+
| 0.1296 | 1.9476 | 4200 | 0.1626 |
|
| 138 |
+
| 0.1299 | 1.9708 | 4250 | 0.1632 |
|
| 139 |
+
| 0.1299 | 1.9940 | 4300 | 0.1605 |
|
| 140 |
+
| 0.1296 | 2.0172 | 4350 | 0.1620 |
|
| 141 |
+
| 0.1302 | 2.0403 | 4400 | 0.1628 |
|
| 142 |
+
| 0.13 | 2.0635 | 4450 | 0.1621 |
|
| 143 |
+
| 0.1296 | 2.0867 | 4500 | 0.1616 |
|
| 144 |
+
| 0.1298 | 2.1099 | 4550 | 0.1613 |
|
| 145 |
+
| 0.1299 | 2.1331 | 4600 | 0.1603 |
|
| 146 |
+
| 0.1299 | 2.1563 | 4650 | 0.1621 |
|
| 147 |
+
| 0.1306 | 2.1795 | 4700 | 0.1614 |
|
| 148 |
+
| 0.1303 | 2.2026 | 4750 | 0.1625 |
|
| 149 |
+
| 0.13 | 2.2258 | 4800 | 0.1624 |
|
| 150 |
+
| 0.1295 | 2.2490 | 4850 | 0.1627 |
|
| 151 |
+
| 0.1299 | 2.2722 | 4900 | 0.1609 |
|
| 152 |
+
| 0.13 | 2.2954 | 4950 | 0.1622 |
|
| 153 |
+
| 0.1311 | 2.3186 | 5000 | 0.1602 |
|
| 154 |
+
| 0.1284 | 2.3418 | 5050 | 0.1616 |
|
| 155 |
+
| 0.13 | 2.3649 | 5100 | 0.1602 |
|
| 156 |
+
| 0.129 | 2.3881 | 5150 | 0.1605 |
|
| 157 |
+
| 0.129 | 2.4113 | 5200 | 0.1606 |
|
| 158 |
+
| 0.1297 | 2.4345 | 5250 | 0.1620 |
|
| 159 |
+
| 0.1293 | 2.4577 | 5300 | 0.1607 |
|
| 160 |
+
| 0.1288 | 2.4809 | 5350 | 0.1615 |
|
| 161 |
+
| 0.1294 | 2.5041 | 5400 | 0.1614 |
|
| 162 |
+
| 0.1285 | 2.5272 | 5450 | 0.1620 |
|
| 163 |
+
| 0.1303 | 2.5504 | 5500 | 0.1618 |
|
| 164 |
+
| 0.1291 | 2.5736 | 5550 | 0.1603 |
|
| 165 |
+
| 0.1298 | 2.5968 | 5600 | 0.1609 |
|
| 166 |
+
| 0.1288 | 2.6200 | 5650 | 0.1604 |
|
| 167 |
+
| 0.129 | 2.6432 | 5700 | 0.1600 |
|
| 168 |
+
| 0.1291 | 2.6664 | 5750 | 0.1597 |
|
| 169 |
+
| 0.1291 | 2.6895 | 5800 | 0.1609 |
|
| 170 |
+
| 0.129 | 2.7127 | 5850 | 0.1611 |
|
| 171 |
+
| 0.13 | 2.7359 | 5900 | 0.1600 |
|
| 172 |
+
| 0.1296 | 2.7591 | 5950 | 0.1603 |
|
| 173 |
+
| 0.1294 | 2.7823 | 6000 | 0.1592 |
|
| 174 |
+
| 0.1283 | 2.8055 | 6050 | 0.1618 |
|
| 175 |
+
| 0.1292 | 2.8287 | 6100 | 0.1612 |
|
| 176 |
+
| 0.128 | 2.8518 | 6150 | 0.1604 |
|
| 177 |
+
| 0.1288 | 2.8750 | 6200 | 0.1611 |
|
| 178 |
+
| 0.1283 | 2.8982 | 6250 | 0.1609 |
|
| 179 |
+
| 0.1292 | 2.9214 | 6300 | 0.1605 |
|
| 180 |
+
| 0.1302 | 2.9446 | 6350 | 0.1602 |
|
| 181 |
+
| 0.1285 | 2.9678 | 6400 | 0.1601 |
|
| 182 |
+
| 0.1286 | 2.9910 | 6450 | 0.1609 |
|
| 183 |
+
| 0.1301 | 3.0141 | 6500 | 0.1602 |
|
| 184 |
+
| 0.1296 | 3.0373 | 6550 | 0.1597 |
|
| 185 |
+
| 0.1291 | 3.0605 | 6600 | 0.1604 |
|
| 186 |
+
| 0.1288 | 3.0837 | 6650 | 0.1595 |
|
| 187 |
+
| 0.129 | 3.1069 | 6700 | 0.1593 |
|
| 188 |
+
| 0.1286 | 3.1301 | 6750 | 0.1600 |
|
| 189 |
+
| 0.1293 | 3.1533 | 6800 | 0.1599 |
|
| 190 |
+
| 0.1289 | 3.1764 | 6850 | 0.1599 |
|
| 191 |
+
| 0.1295 | 3.1996 | 6900 | 0.1601 |
|
| 192 |
+
| 0.1287 | 3.2228 | 6950 | 0.1592 |
|
| 193 |
+
| 0.1286 | 3.2460 | 7000 | 0.1600 |
|
| 194 |
+
| 0.1283 | 3.2692 | 7050 | 0.1598 |
|
| 195 |
+
| 0.1288 | 3.2924 | 7100 | 0.1612 |
|
| 196 |
+
| 0.1298 | 3.3156 | 7150 | 0.1597 |
|
| 197 |
+
| 0.1284 | 3.3387 | 7200 | 0.1605 |
|
| 198 |
+
| 0.1289 | 3.3619 | 7250 | 0.1605 |
|
| 199 |
+
| 0.1289 | 3.3851 | 7300 | 0.1600 |
|
| 200 |
+
| 0.1285 | 3.4083 | 7350 | 0.1605 |
|
| 201 |
+
| 0.1286 | 3.4315 | 7400 | 0.1610 |
|
| 202 |
+
| 0.1278 | 3.4547 | 7450 | 0.1598 |
|
| 203 |
+
| 0.1274 | 3.4779 | 7500 | 0.1598 |
|
| 204 |
+
| 0.1297 | 3.5010 | 7550 | 0.1599 |
|
| 205 |
+
| 0.1288 | 3.5242 | 7600 | 0.1591 |
|
| 206 |
+
| 0.1281 | 3.5474 | 7650 | 0.1598 |
|
| 207 |
+
| 0.1288 | 3.5706 | 7700 | 0.1600 |
|
| 208 |
+
| 0.128 | 3.5938 | 7750 | 0.1594 |
|
| 209 |
+
| 0.1287 | 3.6170 | 7800 | 0.1603 |
|
| 210 |
+
| 0.1291 | 3.6402 | 7850 | 0.1592 |
|
| 211 |
+
| 0.1287 | 3.6633 | 7900 | 0.1596 |
|
| 212 |
+
| 0.1283 | 3.6865 | 7950 | 0.1590 |
|
| 213 |
+
| 0.128 | 3.7097 | 8000 | 0.1584 |
|
| 214 |
+
| 0.1276 | 3.7329 | 8050 | 0.1602 |
|
| 215 |
+
| 0.1287 | 3.7561 | 8100 | 0.1602 |
|
| 216 |
+
| 0.1306 | 3.7793 | 8150 | 0.1595 |
|
| 217 |
+
| 0.1286 | 3.8025 | 8200 | 0.1587 |
|
| 218 |
+
| 0.1292 | 3.8256 | 8250 | 0.1593 |
|
| 219 |
+
| 0.1275 | 3.8488 | 8300 | 0.1590 |
|
| 220 |
+
| 0.1277 | 3.8720 | 8350 | 0.1600 |
|
| 221 |
+
| 0.129 | 3.8952 | 8400 | 0.1602 |
|
| 222 |
+
| 0.1286 | 3.9184 | 8450 | 0.1593 |
|
| 223 |
+
| 0.1281 | 3.9416 | 8500 | 0.1603 |
|
| 224 |
+
| 0.1285 | 3.9648 | 8550 | 0.1591 |
|
| 225 |
+
| 0.1293 | 3.9879 | 8600 | 0.1592 |
|
| 226 |
+
| 0.1283 | 4.0111 | 8650 | 0.1587 |
|
| 227 |
+
| 0.1277 | 4.0343 | 8700 | 0.1598 |
|
| 228 |
+
| 0.1283 | 4.0575 | 8750 | 0.1599 |
|
| 229 |
+
| 0.1288 | 4.0807 | 8800 | 0.1579 |
|
| 230 |
+
| 0.1287 | 4.1039 | 8850 | 0.1588 |
|
| 231 |
+
| 0.1294 | 4.1271 | 8900 | 0.1607 |
|
| 232 |
+
| 0.1277 | 4.1502 | 8950 | 0.1599 |
|
| 233 |
+
| 0.1285 | 4.1734 | 9000 | 0.1595 |
|
| 234 |
+
| 0.1289 | 4.1966 | 9050 | 0.1610 |
|
| 235 |
+
| 0.1289 | 4.2198 | 9100 | 0.1599 |
|
| 236 |
+
| 0.1283 | 4.2430 | 9150 | 0.1589 |
|
| 237 |
+
| 0.1282 | 4.2662 | 9200 | 0.1597 |
|
| 238 |
+
| 0.1286 | 4.2894 | 9250 | 0.1608 |
|
| 239 |
+
| 0.1287 | 4.3125 | 9300 | 0.1608 |
|
| 240 |
+
| 0.1287 | 4.3357 | 9350 | 0.1602 |
|
| 241 |
+
| 0.1286 | 4.3589 | 9400 | 0.1596 |
|
| 242 |
+
| 0.1289 | 4.3821 | 9450 | 0.1598 |
|
| 243 |
+
| 0.1286 | 4.4053 | 9500 | 0.1612 |
|
| 244 |
+
| 0.1281 | 4.4285 | 9550 | 0.1590 |
|
| 245 |
+
| 0.1276 | 4.4517 | 9600 | 0.1588 |
|
| 246 |
+
| 0.1289 | 4.4748 | 9650 | 0.1590 |
|
| 247 |
+
| 0.1284 | 4.4980 | 9700 | 0.1587 |
|
| 248 |
+
| 0.1284 | 4.5212 | 9750 | 0.1597 |
|
| 249 |
+
| 0.1297 | 4.5444 | 9800 | 0.1594 |
|
| 250 |
+
| 0.1276 | 4.5676 | 9850 | 0.1593 |
|
| 251 |
+
| 0.129 | 4.5908 | 9900 | 0.1592 |
|
| 252 |
+
| 0.1285 | 4.6140 | 9950 | 0.1603 |
|
| 253 |
+
| 0.1282 | 4.6371 | 10000 | 0.1601 |
|
| 254 |
+
|
| 255 |
+
|
| 256 |
+
### Framework versions
|
| 257 |
+
|
| 258 |
+
- Transformers 4.40.2
|
| 259 |
+
- Pytorch 2.8.0+cu126
|
| 260 |
+
- Datasets 4.0.0
|
| 261 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.637143519591931,
|
| 3 |
+
"eval_loss": 0.15925397718542236,
|
| 4 |
+
"eval_runtime": 61.5353,
|
| 5 |
+
"eval_samples_per_second": 675.807,
|
| 6 |
+
"eval_steps_per_second": 0.666,
|
| 7 |
+
"perplexity": 1.1726357315864648,
|
| 8 |
+
"total_flos": 2.3231400526217216e+17,
|
| 9 |
+
"train_loss": 0.13326009378433226,
|
| 10 |
+
"train_runtime": 41368.3669,
|
| 11 |
+
"train_samples_per_second": 495.064,
|
| 12 |
+
"train_steps_per_second": 0.242
|
| 13 |
+
}
|
checkpoint-8800/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"GPNRoFormerForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"aux_features_vocab_size": 5,
|
| 7 |
+
"embedding_size": 768,
|
| 8 |
+
"group_tokens": 1,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 1536,
|
| 16 |
+
"model_type": "GPNRoFormer",
|
| 17 |
+
"n_aux_features": 445,
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 2,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"rotary_value": false,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.40.2",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 6
|
| 27 |
+
}
|
checkpoint-8800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3fe36920de849781b2a7f361176544f9d0a1ee3f2ebe62c3627426ee9518250
|
| 3 |
+
size 118212491
|
checkpoint-8800/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0e79347487c61d8217106a3f8f05bbf42d7c2038dab7a7a461077975e6acff9
|
| 3 |
+
size 59497084
|
checkpoint-8800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8637e509d3b3a1df7b3097e625b6e4859dba03dc277087294b9305e0298e9f05
|
| 3 |
+
size 14709
|
checkpoint-8800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b0294dd3bf153ec3af58682fe4ad6efb3b9c01758053c3cd8df76a3af27e88fb
|
| 3 |
+
size 1465
|
checkpoint-8800/trainer_state.json
ADDED
|
@@ -0,0 +1,2661 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.15788726458515429,
|
| 3 |
+
"best_model_checkpoint": "checkpoints/checkpoint-8800",
|
| 4 |
+
"epoch": 4.0806862972408995,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 8800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.023185717597959656,
|
| 13 |
+
"grad_norm": 0.16052097082138062,
|
| 14 |
+
"learning_rate": 5e-05,
|
| 15 |
+
"loss": 0.6225,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.023185717597959656,
|
| 20 |
+
"eval_loss": 0.1987911110084725,
|
| 21 |
+
"eval_runtime": 63.5433,
|
| 22 |
+
"eval_samples_per_second": 654.451,
|
| 23 |
+
"eval_steps_per_second": 0.645,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.04637143519591931,
|
| 28 |
+
"grad_norm": 0.09532159566879272,
|
| 29 |
+
"learning_rate": 0.0001,
|
| 30 |
+
"loss": 0.1508,
|
| 31 |
+
"step": 100
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.04637143519591931,
|
| 35 |
+
"eval_loss": 0.18357936446787168,
|
| 36 |
+
"eval_runtime": 60.9844,
|
| 37 |
+
"eval_samples_per_second": 681.912,
|
| 38 |
+
"eval_steps_per_second": 0.672,
|
| 39 |
+
"step": 100
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06955715279387897,
|
| 43 |
+
"grad_norm": 0.24056212604045868,
|
| 44 |
+
"learning_rate": 9.999370638369377e-05,
|
| 45 |
+
"loss": 0.1449,
|
| 46 |
+
"step": 150
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.06955715279387897,
|
| 50 |
+
"eval_loss": 0.17892658896642444,
|
| 51 |
+
"eval_runtime": 60.7834,
|
| 52 |
+
"eval_samples_per_second": 684.167,
|
| 53 |
+
"eval_steps_per_second": 0.675,
|
| 54 |
+
"step": 150
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09274287039183862,
|
| 58 |
+
"grad_norm": 0.09350813180208206,
|
| 59 |
+
"learning_rate": 9.997482711915927e-05,
|
| 60 |
+
"loss": 0.1421,
|
| 61 |
+
"step": 200
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.09274287039183862,
|
| 65 |
+
"eval_loss": 0.17624869869175752,
|
| 66 |
+
"eval_runtime": 60.3826,
|
| 67 |
+
"eval_samples_per_second": 688.708,
|
| 68 |
+
"eval_steps_per_second": 0.679,
|
| 69 |
+
"step": 200
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.11592858798979828,
|
| 73 |
+
"grad_norm": 0.12230529636144638,
|
| 74 |
+
"learning_rate": 9.99433669591504e-05,
|
| 75 |
+
"loss": 0.141,
|
| 76 |
+
"step": 250
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.11592858798979828,
|
| 80 |
+
"eval_loss": 0.17641382363047173,
|
| 81 |
+
"eval_runtime": 60.4169,
|
| 82 |
+
"eval_samples_per_second": 688.317,
|
| 83 |
+
"eval_steps_per_second": 0.679,
|
| 84 |
+
"step": 250
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.13911430558775795,
|
| 88 |
+
"grad_norm": 0.14592748880386353,
|
| 89 |
+
"learning_rate": 9.989933382359422e-05,
|
| 90 |
+
"loss": 0.1397,
|
| 91 |
+
"step": 300
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.13911430558775795,
|
| 95 |
+
"eval_loss": 0.17552215792639078,
|
| 96 |
+
"eval_runtime": 61.6101,
|
| 97 |
+
"eval_samples_per_second": 674.987,
|
| 98 |
+
"eval_steps_per_second": 0.665,
|
| 99 |
+
"step": 300
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.1623000231857176,
|
| 103 |
+
"grad_norm": 0.10219988226890564,
|
| 104 |
+
"learning_rate": 9.984273879759713e-05,
|
| 105 |
+
"loss": 0.1393,
|
| 106 |
+
"step": 350
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.1623000231857176,
|
| 110 |
+
"eval_loss": 0.17414749172793012,
|
| 111 |
+
"eval_runtime": 61.4962,
|
| 112 |
+
"eval_samples_per_second": 676.237,
|
| 113 |
+
"eval_steps_per_second": 0.667,
|
| 114 |
+
"step": 350
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.18548574078367724,
|
| 118 |
+
"grad_norm": 0.12338168174028397,
|
| 119 |
+
"learning_rate": 9.977359612865423e-05,
|
| 120 |
+
"loss": 0.1388,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.18548574078367724,
|
| 125 |
+
"eval_loss": 0.17378012638412807,
|
| 126 |
+
"eval_runtime": 61.0462,
|
| 127 |
+
"eval_samples_per_second": 681.221,
|
| 128 |
+
"eval_steps_per_second": 0.672,
|
| 129 |
+
"step": 400
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.20867145838163692,
|
| 133 |
+
"grad_norm": 0.09479879587888718,
|
| 134 |
+
"learning_rate": 9.969192322306271e-05,
|
| 135 |
+
"loss": 0.1394,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.20867145838163692,
|
| 140 |
+
"eval_loss": 0.17252362204688398,
|
| 141 |
+
"eval_runtime": 60.963,
|
| 142 |
+
"eval_samples_per_second": 682.151,
|
| 143 |
+
"eval_steps_per_second": 0.673,
|
| 144 |
+
"step": 450
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.23185717597959657,
|
| 148 |
+
"grad_norm": 0.1108623668551445,
|
| 149 |
+
"learning_rate": 9.959774064153977e-05,
|
| 150 |
+
"loss": 0.1383,
|
| 151 |
+
"step": 500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.23185717597959657,
|
| 155 |
+
"eval_loss": 0.17298176916843877,
|
| 156 |
+
"eval_runtime": 60.6546,
|
| 157 |
+
"eval_samples_per_second": 685.62,
|
| 158 |
+
"eval_steps_per_second": 0.676,
|
| 159 |
+
"step": 500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.2550428935775562,
|
| 163 |
+
"grad_norm": 0.0725204199552536,
|
| 164 |
+
"learning_rate": 9.949107209404665e-05,
|
| 165 |
+
"loss": 0.1376,
|
| 166 |
+
"step": 550
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.2550428935775562,
|
| 170 |
+
"eval_loss": 0.17165218165539878,
|
| 171 |
+
"eval_runtime": 59.4888,
|
| 172 |
+
"eval_samples_per_second": 699.056,
|
| 173 |
+
"eval_steps_per_second": 0.689,
|
| 174 |
+
"step": 550
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.2782286111755159,
|
| 178 |
+
"grad_norm": 0.0955963134765625,
|
| 179 |
+
"learning_rate": 9.937194443381972e-05,
|
| 180 |
+
"loss": 0.1372,
|
| 181 |
+
"step": 600
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.2782286111755159,
|
| 185 |
+
"eval_loss": 0.17077083113718278,
|
| 186 |
+
"eval_runtime": 60.6021,
|
| 187 |
+
"eval_samples_per_second": 686.214,
|
| 188 |
+
"eval_steps_per_second": 0.677,
|
| 189 |
+
"step": 600
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.3014143287734755,
|
| 193 |
+
"grad_norm": 0.18736732006072998,
|
| 194 |
+
"learning_rate": 9.924038765061042e-05,
|
| 195 |
+
"loss": 0.1361,
|
| 196 |
+
"step": 650
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.3014143287734755,
|
| 200 |
+
"eval_loss": 0.1727738813183492,
|
| 201 |
+
"eval_runtime": 60.3343,
|
| 202 |
+
"eval_samples_per_second": 689.259,
|
| 203 |
+
"eval_steps_per_second": 0.68,
|
| 204 |
+
"step": 650
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.3246000463714352,
|
| 208 |
+
"grad_norm": 0.09572151303291321,
|
| 209 |
+
"learning_rate": 9.909643486313533e-05,
|
| 210 |
+
"loss": 0.1362,
|
| 211 |
+
"step": 700
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.3246000463714352,
|
| 215 |
+
"eval_loss": 0.17145407115151273,
|
| 216 |
+
"eval_runtime": 60.2732,
|
| 217 |
+
"eval_samples_per_second": 689.959,
|
| 218 |
+
"eval_steps_per_second": 0.68,
|
| 219 |
+
"step": 700
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.34778576396939487,
|
| 223 |
+
"grad_norm": 0.07214252650737762,
|
| 224 |
+
"learning_rate": 9.894012231073894e-05,
|
| 225 |
+
"loss": 0.1364,
|
| 226 |
+
"step": 750
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.34778576396939487,
|
| 230 |
+
"eval_loss": 0.17133199408489355,
|
| 231 |
+
"eval_runtime": 60.0148,
|
| 232 |
+
"eval_samples_per_second": 692.929,
|
| 233 |
+
"eval_steps_per_second": 0.683,
|
| 234 |
+
"step": 750
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.3709714815673545,
|
| 238 |
+
"grad_norm": 0.18224318325519562,
|
| 239 |
+
"learning_rate": 9.877148934427037e-05,
|
| 240 |
+
"loss": 0.1356,
|
| 241 |
+
"step": 800
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3709714815673545,
|
| 245 |
+
"eval_loss": 0.16949569222888886,
|
| 246 |
+
"eval_runtime": 60.1491,
|
| 247 |
+
"eval_samples_per_second": 691.382,
|
| 248 |
+
"eval_steps_per_second": 0.682,
|
| 249 |
+
"step": 800
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.39415719916531416,
|
| 253 |
+
"grad_norm": 0.06306415796279907,
|
| 254 |
+
"learning_rate": 9.859057841617709e-05,
|
| 255 |
+
"loss": 0.1353,
|
| 256 |
+
"step": 850
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.39415719916531416,
|
| 260 |
+
"eval_loss": 0.1686952690798172,
|
| 261 |
+
"eval_runtime": 60.6447,
|
| 262 |
+
"eval_samples_per_second": 685.731,
|
| 263 |
+
"eval_steps_per_second": 0.676,
|
| 264 |
+
"step": 850
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.41734291676327384,
|
| 268 |
+
"grad_norm": 0.10090287029743195,
|
| 269 |
+
"learning_rate": 9.839743506981782e-05,
|
| 270 |
+
"loss": 0.1361,
|
| 271 |
+
"step": 900
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.41734291676327384,
|
| 275 |
+
"eval_loss": 0.17026100034088926,
|
| 276 |
+
"eval_runtime": 61.5224,
|
| 277 |
+
"eval_samples_per_second": 675.949,
|
| 278 |
+
"eval_steps_per_second": 0.666,
|
| 279 |
+
"step": 900
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.44052863436123346,
|
| 283 |
+
"grad_norm": 0.10061236470937729,
|
| 284 |
+
"learning_rate": 9.819210792799712e-05,
|
| 285 |
+
"loss": 0.1354,
|
| 286 |
+
"step": 950
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.44052863436123346,
|
| 290 |
+
"eval_loss": 0.16971544565694113,
|
| 291 |
+
"eval_runtime": 60.488,
|
| 292 |
+
"eval_samples_per_second": 687.508,
|
| 293 |
+
"eval_steps_per_second": 0.678,
|
| 294 |
+
"step": 950
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.46371435195919314,
|
| 298 |
+
"grad_norm": 0.06525534391403198,
|
| 299 |
+
"learning_rate": 9.797464868072488e-05,
|
| 300 |
+
"loss": 0.1352,
|
| 301 |
+
"step": 1000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.46371435195919314,
|
| 305 |
+
"eval_loss": 0.16946903195393553,
|
| 306 |
+
"eval_runtime": 61.3558,
|
| 307 |
+
"eval_samples_per_second": 677.784,
|
| 308 |
+
"eval_steps_per_second": 0.668,
|
| 309 |
+
"step": 1000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.4869000695571528,
|
| 313 |
+
"grad_norm": 0.06269507855176926,
|
| 314 |
+
"learning_rate": 9.77451120722037e-05,
|
| 315 |
+
"loss": 0.1335,
|
| 316 |
+
"step": 1050
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.4869000695571528,
|
| 320 |
+
"eval_loss": 0.16825352947444114,
|
| 321 |
+
"eval_runtime": 60.1971,
|
| 322 |
+
"eval_samples_per_second": 690.831,
|
| 323 |
+
"eval_steps_per_second": 0.681,
|
| 324 |
+
"step": 1050
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5100857871551124,
|
| 328 |
+
"grad_norm": 0.08187470585107803,
|
| 329 |
+
"learning_rate": 9.750355588704727e-05,
|
| 330 |
+
"loss": 0.1327,
|
| 331 |
+
"step": 1100
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5100857871551124,
|
| 335 |
+
"eval_loss": 0.16861737282523587,
|
| 336 |
+
"eval_runtime": 59.5715,
|
| 337 |
+
"eval_samples_per_second": 698.085,
|
| 338 |
+
"eval_steps_per_second": 0.688,
|
| 339 |
+
"step": 1100
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5332715047530721,
|
| 343 |
+
"grad_norm": 0.06607680767774582,
|
| 344 |
+
"learning_rate": 9.725004093573342e-05,
|
| 345 |
+
"loss": 0.1337,
|
| 346 |
+
"step": 1150
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5332715047530721,
|
| 350 |
+
"eval_loss": 0.1692070748762034,
|
| 351 |
+
"eval_runtime": 60.0637,
|
| 352 |
+
"eval_samples_per_second": 692.364,
|
| 353 |
+
"eval_steps_per_second": 0.683,
|
| 354 |
+
"step": 1150
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.5564572223510318,
|
| 358 |
+
"grad_norm": 0.09759815782308578,
|
| 359 |
+
"learning_rate": 9.698463103929542e-05,
|
| 360 |
+
"loss": 0.134,
|
| 361 |
+
"step": 1200
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5564572223510318,
|
| 365 |
+
"eval_loss": 0.16649228385381692,
|
| 366 |
+
"eval_runtime": 60.4305,
|
| 367 |
+
"eval_samples_per_second": 688.162,
|
| 368 |
+
"eval_steps_per_second": 0.678,
|
| 369 |
+
"step": 1200
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.5796429399489914,
|
| 373 |
+
"grad_norm": 0.10353852063417435,
|
| 374 |
+
"learning_rate": 9.670739301325534e-05,
|
| 375 |
+
"loss": 0.1341,
|
| 376 |
+
"step": 1250
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.5796429399489914,
|
| 380 |
+
"eval_loss": 0.16802514322459206,
|
| 381 |
+
"eval_runtime": 60.0955,
|
| 382 |
+
"eval_samples_per_second": 691.999,
|
| 383 |
+
"eval_steps_per_second": 0.682,
|
| 384 |
+
"step": 1250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.602828657546951,
|
| 388 |
+
"grad_norm": 0.11834366619586945,
|
| 389 |
+
"learning_rate": 9.641839665080363e-05,
|
| 390 |
+
"loss": 0.1347,
|
| 391 |
+
"step": 1300
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.602828657546951,
|
| 395 |
+
"eval_loss": 0.1672302417427292,
|
| 396 |
+
"eval_runtime": 60.3484,
|
| 397 |
+
"eval_samples_per_second": 689.098,
|
| 398 |
+
"eval_steps_per_second": 0.679,
|
| 399 |
+
"step": 1300
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.6260143751449108,
|
| 403 |
+
"grad_norm": 0.06963012367486954,
|
| 404 |
+
"learning_rate": 9.611771470522908e-05,
|
| 405 |
+
"loss": 0.1335,
|
| 406 |
+
"step": 1350
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.6260143751449108,
|
| 410 |
+
"eval_loss": 0.16607839684977216,
|
| 411 |
+
"eval_runtime": 60.3308,
|
| 412 |
+
"eval_samples_per_second": 689.3,
|
| 413 |
+
"eval_steps_per_second": 0.68,
|
| 414 |
+
"step": 1350
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.6492000927428704,
|
| 418 |
+
"grad_norm": 0.06842990219593048,
|
| 419 |
+
"learning_rate": 9.580542287160348e-05,
|
| 420 |
+
"loss": 0.1338,
|
| 421 |
+
"step": 1400
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.6492000927428704,
|
| 425 |
+
"eval_loss": 0.16628812684035693,
|
| 426 |
+
"eval_runtime": 59.9335,
|
| 427 |
+
"eval_samples_per_second": 693.87,
|
| 428 |
+
"eval_steps_per_second": 0.684,
|
| 429 |
+
"step": 1400
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.67238581034083,
|
| 433 |
+
"grad_norm": 0.07053674757480621,
|
| 434 |
+
"learning_rate": 9.548159976772592e-05,
|
| 435 |
+
"loss": 0.1335,
|
| 436 |
+
"step": 1450
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.67238581034083,
|
| 440 |
+
"eval_loss": 0.16696060882262428,
|
| 441 |
+
"eval_runtime": 59.8079,
|
| 442 |
+
"eval_samples_per_second": 695.326,
|
| 443 |
+
"eval_steps_per_second": 0.686,
|
| 444 |
+
"step": 1450
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6955715279387897,
|
| 448 |
+
"grad_norm": 0.09175281971693039,
|
| 449 |
+
"learning_rate": 9.514632691433107e-05,
|
| 450 |
+
"loss": 0.1332,
|
| 451 |
+
"step": 1500
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6955715279387897,
|
| 455 |
+
"eval_loss": 0.16521949465081834,
|
| 456 |
+
"eval_runtime": 60.1856,
|
| 457 |
+
"eval_samples_per_second": 690.963,
|
| 458 |
+
"eval_steps_per_second": 0.681,
|
| 459 |
+
"step": 1500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.7187572455367494,
|
| 463 |
+
"grad_norm": 0.05836635082960129,
|
| 464 |
+
"learning_rate": 9.479968871456679e-05,
|
| 465 |
+
"loss": 0.1336,
|
| 466 |
+
"step": 1550
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.7187572455367494,
|
| 470 |
+
"eval_loss": 0.16626366255041727,
|
| 471 |
+
"eval_runtime": 60.6256,
|
| 472 |
+
"eval_samples_per_second": 685.948,
|
| 473 |
+
"eval_steps_per_second": 0.676,
|
| 474 |
+
"step": 1550
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.741942963134709,
|
| 478 |
+
"grad_norm": 0.07249301671981812,
|
| 479 |
+
"learning_rate": 9.444177243274618e-05,
|
| 480 |
+
"loss": 0.133,
|
| 481 |
+
"step": 1600
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.741942963134709,
|
| 485 |
+
"eval_loss": 0.1655649439629329,
|
| 486 |
+
"eval_runtime": 60.2447,
|
| 487 |
+
"eval_samples_per_second": 690.285,
|
| 488 |
+
"eval_steps_per_second": 0.681,
|
| 489 |
+
"step": 1600
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.7651286807326687,
|
| 493 |
+
"grad_norm": 0.07509302347898483,
|
| 494 |
+
"learning_rate": 9.407266817237911e-05,
|
| 495 |
+
"loss": 0.1332,
|
| 496 |
+
"step": 1650
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.7651286807326687,
|
| 500 |
+
"eval_loss": 0.16605371203296967,
|
| 501 |
+
"eval_runtime": 59.8196,
|
| 502 |
+
"eval_samples_per_second": 695.191,
|
| 503 |
+
"eval_steps_per_second": 0.685,
|
| 504 |
+
"step": 1650
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.7883143983306283,
|
| 508 |
+
"grad_norm": 0.07540406286716461,
|
| 509 |
+
"learning_rate": 9.369246885348926e-05,
|
| 510 |
+
"loss": 0.1327,
|
| 511 |
+
"step": 1700
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.7883143983306283,
|
| 515 |
+
"eval_loss": 0.16555590021301406,
|
| 516 |
+
"eval_runtime": 60.4119,
|
| 517 |
+
"eval_samples_per_second": 688.374,
|
| 518 |
+
"eval_steps_per_second": 0.679,
|
| 519 |
+
"step": 1700
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.811500115928588,
|
| 523 |
+
"grad_norm": 0.06061087176203728,
|
| 524 |
+
"learning_rate": 9.330127018922194e-05,
|
| 525 |
+
"loss": 0.1318,
|
| 526 |
+
"step": 1750
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.811500115928588,
|
| 530 |
+
"eval_loss": 0.16623179673527624,
|
| 531 |
+
"eval_runtime": 59.7807,
|
| 532 |
+
"eval_samples_per_second": 695.643,
|
| 533 |
+
"eval_steps_per_second": 0.686,
|
| 534 |
+
"step": 1750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.8346858335265477,
|
| 538 |
+
"grad_norm": 0.05577518790960312,
|
| 539 |
+
"learning_rate": 9.289917066174886e-05,
|
| 540 |
+
"loss": 0.1319,
|
| 541 |
+
"step": 1800
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.8346858335265477,
|
| 545 |
+
"eval_loss": 0.16519989030959317,
|
| 546 |
+
"eval_runtime": 60.1508,
|
| 547 |
+
"eval_samples_per_second": 691.363,
|
| 548 |
+
"eval_steps_per_second": 0.682,
|
| 549 |
+
"step": 1800
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8578715511245073,
|
| 553 |
+
"grad_norm": 0.06929640471935272,
|
| 554 |
+
"learning_rate": 9.248627149747573e-05,
|
| 555 |
+
"loss": 0.1337,
|
| 556 |
+
"step": 1850
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.8578715511245073,
|
| 560 |
+
"eval_loss": 0.16394849849125304,
|
| 561 |
+
"eval_runtime": 60.1044,
|
| 562 |
+
"eval_samples_per_second": 691.896,
|
| 563 |
+
"eval_steps_per_second": 0.682,
|
| 564 |
+
"step": 1850
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.8810572687224669,
|
| 568 |
+
"grad_norm": 0.07941466569900513,
|
| 569 |
+
"learning_rate": 9.206267664155907e-05,
|
| 570 |
+
"loss": 0.1324,
|
| 571 |
+
"step": 1900
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.8810572687224669,
|
| 575 |
+
"eval_loss": 0.1648257591054525,
|
| 576 |
+
"eval_runtime": 59.9818,
|
| 577 |
+
"eval_samples_per_second": 693.31,
|
| 578 |
+
"eval_steps_per_second": 0.684,
|
| 579 |
+
"step": 1900
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.9042429863204267,
|
| 583 |
+
"grad_norm": 0.09700328856706619,
|
| 584 |
+
"learning_rate": 9.162849273173857e-05,
|
| 585 |
+
"loss": 0.1334,
|
| 586 |
+
"step": 1950
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.9042429863204267,
|
| 590 |
+
"eval_loss": 0.16508159820082235,
|
| 591 |
+
"eval_runtime": 60.0956,
|
| 592 |
+
"eval_samples_per_second": 691.997,
|
| 593 |
+
"eval_steps_per_second": 0.682,
|
| 594 |
+
"step": 1950
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.9274287039183863,
|
| 598 |
+
"grad_norm": 0.09397923946380615,
|
| 599 |
+
"learning_rate": 9.118382907149165e-05,
|
| 600 |
+
"loss": 0.1317,
|
| 601 |
+
"step": 2000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.9274287039183863,
|
| 605 |
+
"eval_loss": 0.16377645660046958,
|
| 606 |
+
"eval_runtime": 60.471,
|
| 607 |
+
"eval_samples_per_second": 687.701,
|
| 608 |
+
"eval_steps_per_second": 0.678,
|
| 609 |
+
"step": 2000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.9506144215163459,
|
| 613 |
+
"grad_norm": 0.08097202330827713,
|
| 614 |
+
"learning_rate": 9.072879760251679e-05,
|
| 615 |
+
"loss": 0.1324,
|
| 616 |
+
"step": 2050
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.9506144215163459,
|
| 620 |
+
"eval_loss": 0.16491611914973717,
|
| 621 |
+
"eval_runtime": 60.6678,
|
| 622 |
+
"eval_samples_per_second": 685.471,
|
| 623 |
+
"eval_steps_per_second": 0.676,
|
| 624 |
+
"step": 2050
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.9738001391143056,
|
| 628 |
+
"grad_norm": 0.08455361425876617,
|
| 629 |
+
"learning_rate": 9.026351287655294e-05,
|
| 630 |
+
"loss": 0.1326,
|
| 631 |
+
"step": 2100
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.9738001391143056,
|
| 635 |
+
"eval_loss": 0.16602741997032858,
|
| 636 |
+
"eval_runtime": 60.5593,
|
| 637 |
+
"eval_samples_per_second": 686.698,
|
| 638 |
+
"eval_steps_per_second": 0.677,
|
| 639 |
+
"step": 2100
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9969858567122653,
|
| 643 |
+
"grad_norm": 0.056316621601581573,
|
| 644 |
+
"learning_rate": 8.978809202654162e-05,
|
| 645 |
+
"loss": 0.1326,
|
| 646 |
+
"step": 2150
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.9969858567122653,
|
| 650 |
+
"eval_loss": 0.1640188218462461,
|
| 651 |
+
"eval_runtime": 60.9228,
|
| 652 |
+
"eval_samples_per_second": 682.602,
|
| 653 |
+
"eval_steps_per_second": 0.673,
|
| 654 |
+
"step": 2150
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.0201715743102249,
|
| 658 |
+
"grad_norm": 0.06686601787805557,
|
| 659 |
+
"learning_rate": 8.930265473713938e-05,
|
| 660 |
+
"loss": 0.132,
|
| 661 |
+
"step": 2200
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.0201715743102249,
|
| 665 |
+
"eval_loss": 0.1652621257294944,
|
| 666 |
+
"eval_runtime": 60.883,
|
| 667 |
+
"eval_samples_per_second": 683.048,
|
| 668 |
+
"eval_steps_per_second": 0.673,
|
| 669 |
+
"step": 2200
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.0433572919081846,
|
| 673 |
+
"grad_norm": 0.040202509611845016,
|
| 674 |
+
"learning_rate": 8.880732321458784e-05,
|
| 675 |
+
"loss": 0.1319,
|
| 676 |
+
"step": 2250
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.0433572919081846,
|
| 680 |
+
"eval_loss": 0.1655291575008717,
|
| 681 |
+
"eval_runtime": 60.3109,
|
| 682 |
+
"eval_samples_per_second": 689.527,
|
| 683 |
+
"eval_steps_per_second": 0.68,
|
| 684 |
+
"step": 2250
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.0665430095061441,
|
| 688 |
+
"grad_norm": 0.0656428411602974,
|
| 689 |
+
"learning_rate": 8.83022221559489e-05,
|
| 690 |
+
"loss": 0.1326,
|
| 691 |
+
"step": 2300
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 1.0665430095061441,
|
| 695 |
+
"eval_loss": 0.16431572407087935,
|
| 696 |
+
"eval_runtime": 60.1036,
|
| 697 |
+
"eval_samples_per_second": 691.906,
|
| 698 |
+
"eval_steps_per_second": 0.682,
|
| 699 |
+
"step": 2300
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.0897287271041038,
|
| 703 |
+
"grad_norm": 0.06945247948169708,
|
| 704 |
+
"learning_rate": 8.778747871771292e-05,
|
| 705 |
+
"loss": 0.1321,
|
| 706 |
+
"step": 2350
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 1.0897287271041038,
|
| 710 |
+
"eval_loss": 0.16585482329987242,
|
| 711 |
+
"eval_runtime": 60.6263,
|
| 712 |
+
"eval_samples_per_second": 685.94,
|
| 713 |
+
"eval_steps_per_second": 0.676,
|
| 714 |
+
"step": 2350
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.1129144447020636,
|
| 718 |
+
"grad_norm": 0.0523492731153965,
|
| 719 |
+
"learning_rate": 8.726322248378775e-05,
|
| 720 |
+
"loss": 0.1317,
|
| 721 |
+
"step": 2400
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 1.1129144447020636,
|
| 725 |
+
"eval_loss": 0.16438524923736036,
|
| 726 |
+
"eval_runtime": 60.317,
|
| 727 |
+
"eval_samples_per_second": 689.457,
|
| 728 |
+
"eval_steps_per_second": 0.68,
|
| 729 |
+
"step": 2400
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.136100162300023,
|
| 733 |
+
"grad_norm": 0.07777334004640579,
|
| 734 |
+
"learning_rate": 8.672958543287666e-05,
|
| 735 |
+
"loss": 0.1322,
|
| 736 |
+
"step": 2450
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.136100162300023,
|
| 740 |
+
"eval_loss": 0.16509696053644565,
|
| 741 |
+
"eval_runtime": 60.26,
|
| 742 |
+
"eval_samples_per_second": 690.109,
|
| 743 |
+
"eval_steps_per_second": 0.68,
|
| 744 |
+
"step": 2450
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.1592858798979828,
|
| 748 |
+
"grad_norm": 0.06430637836456299,
|
| 749 |
+
"learning_rate": 8.618670190525352e-05,
|
| 750 |
+
"loss": 0.1325,
|
| 751 |
+
"step": 2500
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.1592858798979828,
|
| 755 |
+
"eval_loss": 0.1639541608445008,
|
| 756 |
+
"eval_runtime": 60.5314,
|
| 757 |
+
"eval_samples_per_second": 687.015,
|
| 758 |
+
"eval_steps_per_second": 0.677,
|
| 759 |
+
"step": 2500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.1824715974959426,
|
| 763 |
+
"grad_norm": 0.11194106936454773,
|
| 764 |
+
"learning_rate": 8.563470856894316e-05,
|
| 765 |
+
"loss": 0.1311,
|
| 766 |
+
"step": 2550
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.1824715974959426,
|
| 770 |
+
"eval_loss": 0.16260699934317355,
|
| 771 |
+
"eval_runtime": 60.3659,
|
| 772 |
+
"eval_samples_per_second": 688.899,
|
| 773 |
+
"eval_steps_per_second": 0.679,
|
| 774 |
+
"step": 2550
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.205657315093902,
|
| 778 |
+
"grad_norm": 0.06165901944041252,
|
| 779 |
+
"learning_rate": 8.507374438531607e-05,
|
| 780 |
+
"loss": 0.1323,
|
| 781 |
+
"step": 2600
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.205657315093902,
|
| 785 |
+
"eval_loss": 0.1626319663130242,
|
| 786 |
+
"eval_runtime": 59.9516,
|
| 787 |
+
"eval_samples_per_second": 693.66,
|
| 788 |
+
"eval_steps_per_second": 0.684,
|
| 789 |
+
"step": 2600
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.2288430326918618,
|
| 793 |
+
"grad_norm": 0.10654885321855545,
|
| 794 |
+
"learning_rate": 8.450395057410561e-05,
|
| 795 |
+
"loss": 0.1316,
|
| 796 |
+
"step": 2650
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.2288430326918618,
|
| 800 |
+
"eval_loss": 0.16393000041041636,
|
| 801 |
+
"eval_runtime": 59.576,
|
| 802 |
+
"eval_samples_per_second": 698.032,
|
| 803 |
+
"eval_steps_per_second": 0.688,
|
| 804 |
+
"step": 2650
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.2520287502898215,
|
| 808 |
+
"grad_norm": 0.04848140478134155,
|
| 809 |
+
"learning_rate": 8.392547057785661e-05,
|
| 810 |
+
"loss": 0.1314,
|
| 811 |
+
"step": 2700
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.2520287502898215,
|
| 815 |
+
"eval_loss": 0.16348152455768114,
|
| 816 |
+
"eval_runtime": 60.098,
|
| 817 |
+
"eval_samples_per_second": 691.97,
|
| 818 |
+
"eval_steps_per_second": 0.682,
|
| 819 |
+
"step": 2700
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.275214467887781,
|
| 823 |
+
"grad_norm": 0.0573604516685009,
|
| 824 |
+
"learning_rate": 8.333845002581458e-05,
|
| 825 |
+
"loss": 0.1314,
|
| 826 |
+
"step": 2750
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.275214467887781,
|
| 830 |
+
"eval_loss": 0.16364089140116167,
|
| 831 |
+
"eval_runtime": 60.1364,
|
| 832 |
+
"eval_samples_per_second": 691.528,
|
| 833 |
+
"eval_steps_per_second": 0.682,
|
| 834 |
+
"step": 2750
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.2984001854857408,
|
| 838 |
+
"grad_norm": 0.053159259259700775,
|
| 839 |
+
"learning_rate": 8.274303669726426e-05,
|
| 840 |
+
"loss": 0.131,
|
| 841 |
+
"step": 2800
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.2984001854857408,
|
| 845 |
+
"eval_loss": 0.16257415365129801,
|
| 846 |
+
"eval_runtime": 60.0025,
|
| 847 |
+
"eval_samples_per_second": 693.071,
|
| 848 |
+
"eval_steps_per_second": 0.683,
|
| 849 |
+
"step": 2800
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.3215859030837005,
|
| 853 |
+
"grad_norm": 0.09136148542165756,
|
| 854 |
+
"learning_rate": 8.213938048432697e-05,
|
| 855 |
+
"loss": 0.1313,
|
| 856 |
+
"step": 2850
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.3215859030837005,
|
| 860 |
+
"eval_loss": 0.16324665471619784,
|
| 861 |
+
"eval_runtime": 59.8429,
|
| 862 |
+
"eval_samples_per_second": 694.92,
|
| 863 |
+
"eval_steps_per_second": 0.685,
|
| 864 |
+
"step": 2850
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.34477162068166,
|
| 868 |
+
"grad_norm": 0.05825324356555939,
|
| 869 |
+
"learning_rate": 8.152763335422613e-05,
|
| 870 |
+
"loss": 0.1312,
|
| 871 |
+
"step": 2900
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.34477162068166,
|
| 875 |
+
"eval_loss": 0.16367374608121235,
|
| 876 |
+
"eval_runtime": 60.219,
|
| 877 |
+
"eval_samples_per_second": 690.579,
|
| 878 |
+
"eval_steps_per_second": 0.681,
|
| 879 |
+
"step": 2900
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.3679573382796197,
|
| 883 |
+
"grad_norm": 0.06379790604114532,
|
| 884 |
+
"learning_rate": 8.090794931103026e-05,
|
| 885 |
+
"loss": 0.1317,
|
| 886 |
+
"step": 2950
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.3679573382796197,
|
| 890 |
+
"eval_loss": 0.16400733758786312,
|
| 891 |
+
"eval_runtime": 59.9641,
|
| 892 |
+
"eval_samples_per_second": 693.515,
|
| 893 |
+
"eval_steps_per_second": 0.684,
|
| 894 |
+
"step": 2950
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.3911430558775795,
|
| 898 |
+
"grad_norm": 0.05361103266477585,
|
| 899 |
+
"learning_rate": 8.028048435688333e-05,
|
| 900 |
+
"loss": 0.1311,
|
| 901 |
+
"step": 3000
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 1.3911430558775795,
|
| 905 |
+
"eval_loss": 0.16210626991928834,
|
| 906 |
+
"eval_runtime": 59.5858,
|
| 907 |
+
"eval_samples_per_second": 697.919,
|
| 908 |
+
"eval_steps_per_second": 0.688,
|
| 909 |
+
"step": 3000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.414328773475539,
|
| 913 |
+
"grad_norm": 0.04593402519822121,
|
| 914 |
+
"learning_rate": 7.964539645273204e-05,
|
| 915 |
+
"loss": 0.1304,
|
| 916 |
+
"step": 3050
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 1.414328773475539,
|
| 920 |
+
"eval_loss": 0.163067463275087,
|
| 921 |
+
"eval_runtime": 60.098,
|
| 922 |
+
"eval_samples_per_second": 691.97,
|
| 923 |
+
"eval_steps_per_second": 0.682,
|
| 924 |
+
"step": 3050
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.4375144910734987,
|
| 928 |
+
"grad_norm": 0.057480327785015106,
|
| 929 |
+
"learning_rate": 7.900284547855991e-05,
|
| 930 |
+
"loss": 0.1307,
|
| 931 |
+
"step": 3100
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 1.4375144910734987,
|
| 935 |
+
"eval_loss": 0.16243572043734797,
|
| 936 |
+
"eval_runtime": 59.5674,
|
| 937 |
+
"eval_samples_per_second": 698.133,
|
| 938 |
+
"eval_steps_per_second": 0.688,
|
| 939 |
+
"step": 3100
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.4607002086714584,
|
| 943 |
+
"grad_norm": 0.08223798871040344,
|
| 944 |
+
"learning_rate": 7.835299319313853e-05,
|
| 945 |
+
"loss": 0.1315,
|
| 946 |
+
"step": 3150
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.4607002086714584,
|
| 950 |
+
"eval_loss": 0.1641944734489707,
|
| 951 |
+
"eval_runtime": 59.5423,
|
| 952 |
+
"eval_samples_per_second": 698.428,
|
| 953 |
+
"eval_steps_per_second": 0.689,
|
| 954 |
+
"step": 3150
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.483885926269418,
|
| 958 |
+
"grad_norm": 0.09742949903011322,
|
| 959 |
+
"learning_rate": 7.769600319330552e-05,
|
| 960 |
+
"loss": 0.1303,
|
| 961 |
+
"step": 3200
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.483885926269418,
|
| 965 |
+
"eval_loss": 0.16355698856626613,
|
| 966 |
+
"eval_runtime": 60.1946,
|
| 967 |
+
"eval_samples_per_second": 690.859,
|
| 968 |
+
"eval_steps_per_second": 0.681,
|
| 969 |
+
"step": 3200
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.5070716438673777,
|
| 973 |
+
"grad_norm": 0.06401767581701279,
|
| 974 |
+
"learning_rate": 7.703204087277988e-05,
|
| 975 |
+
"loss": 0.1315,
|
| 976 |
+
"step": 3250
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.5070716438673777,
|
| 980 |
+
"eval_loss": 0.16215006705140952,
|
| 981 |
+
"eval_runtime": 59.7822,
|
| 982 |
+
"eval_samples_per_second": 695.625,
|
| 983 |
+
"eval_steps_per_second": 0.686,
|
| 984 |
+
"step": 3250
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.5302573614653374,
|
| 988 |
+
"grad_norm": 0.07916898280382156,
|
| 989 |
+
"learning_rate": 7.636127338052512e-05,
|
| 990 |
+
"loss": 0.1315,
|
| 991 |
+
"step": 3300
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 1.5302573614653374,
|
| 995 |
+
"eval_loss": 0.16288597734760557,
|
| 996 |
+
"eval_runtime": 59.2757,
|
| 997 |
+
"eval_samples_per_second": 701.57,
|
| 998 |
+
"eval_steps_per_second": 0.692,
|
| 999 |
+
"step": 3300
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.553443079063297,
|
| 1003 |
+
"grad_norm": 0.06549016386270523,
|
| 1004 |
+
"learning_rate": 7.568386957867033e-05,
|
| 1005 |
+
"loss": 0.1303,
|
| 1006 |
+
"step": 3350
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.553443079063297,
|
| 1010 |
+
"eval_loss": 0.16416664097655873,
|
| 1011 |
+
"eval_runtime": 59.84,
|
| 1012 |
+
"eval_samples_per_second": 694.953,
|
| 1013 |
+
"eval_steps_per_second": 0.685,
|
| 1014 |
+
"step": 3350
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 1.5766287966612567,
|
| 1018 |
+
"grad_norm": 0.0709395632147789,
|
| 1019 |
+
"learning_rate": 7.500000000000001e-05,
|
| 1020 |
+
"loss": 0.1309,
|
| 1021 |
+
"step": 3400
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.5766287966612567,
|
| 1025 |
+
"eval_loss": 0.16179194486424098,
|
| 1026 |
+
"eval_runtime": 59.8634,
|
| 1027 |
+
"eval_samples_per_second": 694.682,
|
| 1028 |
+
"eval_steps_per_second": 0.685,
|
| 1029 |
+
"step": 3400
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 1.5998145142592164,
|
| 1033 |
+
"grad_norm": 0.05671363323926926,
|
| 1034 |
+
"learning_rate": 7.430983680502344e-05,
|
| 1035 |
+
"loss": 0.1307,
|
| 1036 |
+
"step": 3450
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 1.5998145142592164,
|
| 1040 |
+
"eval_loss": 0.16309191886303373,
|
| 1041 |
+
"eval_runtime": 59.618,
|
| 1042 |
+
"eval_samples_per_second": 697.541,
|
| 1043 |
+
"eval_steps_per_second": 0.688,
|
| 1044 |
+
"step": 3450
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.623000231857176,
|
| 1048 |
+
"grad_norm": 0.04889162629842758,
|
| 1049 |
+
"learning_rate": 7.361355373863414e-05,
|
| 1050 |
+
"loss": 0.1314,
|
| 1051 |
+
"step": 3500
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.623000231857176,
|
| 1055 |
+
"eval_loss": 0.16290782983414598,
|
| 1056 |
+
"eval_runtime": 60.3904,
|
| 1057 |
+
"eval_samples_per_second": 688.619,
|
| 1058 |
+
"eval_steps_per_second": 0.679,
|
| 1059 |
+
"step": 3500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.6461859494551356,
|
| 1063 |
+
"grad_norm": 0.0970933735370636,
|
| 1064 |
+
"learning_rate": 7.291132608637052e-05,
|
| 1065 |
+
"loss": 0.1314,
|
| 1066 |
+
"step": 3550
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.6461859494551356,
|
| 1070 |
+
"eval_loss": 0.16278222993823557,
|
| 1071 |
+
"eval_runtime": 59.8666,
|
| 1072 |
+
"eval_samples_per_second": 694.644,
|
| 1073 |
+
"eval_steps_per_second": 0.685,
|
| 1074 |
+
"step": 3550
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.6693716670530954,
|
| 1078 |
+
"grad_norm": 0.056557025760412216,
|
| 1079 |
+
"learning_rate": 7.220333063028872e-05,
|
| 1080 |
+
"loss": 0.1312,
|
| 1081 |
+
"step": 3600
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.6693716670530954,
|
| 1085 |
+
"eval_loss": 0.16313205291311117,
|
| 1086 |
+
"eval_runtime": 60.0092,
|
| 1087 |
+
"eval_samples_per_second": 692.993,
|
| 1088 |
+
"eval_steps_per_second": 0.683,
|
| 1089 |
+
"step": 3600
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.6925573846510549,
|
| 1093 |
+
"grad_norm": 0.04870522394776344,
|
| 1094 |
+
"learning_rate": 7.148974560445859e-05,
|
| 1095 |
+
"loss": 0.1299,
|
| 1096 |
+
"step": 3650
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.6925573846510549,
|
| 1100 |
+
"eval_loss": 0.1617941082289122,
|
| 1101 |
+
"eval_runtime": 60.1721,
|
| 1102 |
+
"eval_samples_per_second": 691.117,
|
| 1103 |
+
"eval_steps_per_second": 0.681,
|
| 1104 |
+
"step": 3650
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.7157431022490146,
|
| 1108 |
+
"grad_norm": 0.0681833028793335,
|
| 1109 |
+
"learning_rate": 7.077075065009433e-05,
|
| 1110 |
+
"loss": 0.1304,
|
| 1111 |
+
"step": 3700
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 1.7157431022490146,
|
| 1115 |
+
"eval_loss": 0.16243406602519425,
|
| 1116 |
+
"eval_runtime": 59.3626,
|
| 1117 |
+
"eval_samples_per_second": 700.542,
|
| 1118 |
+
"eval_steps_per_second": 0.691,
|
| 1119 |
+
"step": 3700
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 1.7389288198469743,
|
| 1123 |
+
"grad_norm": 0.06506156921386719,
|
| 1124 |
+
"learning_rate": 7.004652677033068e-05,
|
| 1125 |
+
"loss": 0.1299,
|
| 1126 |
+
"step": 3750
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 1.7389288198469743,
|
| 1130 |
+
"eval_loss": 0.16324780134312317,
|
| 1131 |
+
"eval_runtime": 59.6022,
|
| 1132 |
+
"eval_samples_per_second": 697.726,
|
| 1133 |
+
"eval_steps_per_second": 0.688,
|
| 1134 |
+
"step": 3750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 1.7621145374449338,
|
| 1138 |
+
"grad_norm": 0.06188170611858368,
|
| 1139 |
+
"learning_rate": 6.931725628465643e-05,
|
| 1140 |
+
"loss": 0.1309,
|
| 1141 |
+
"step": 3800
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 1.7621145374449338,
|
| 1145 |
+
"eval_loss": 0.1623115342294882,
|
| 1146 |
+
"eval_runtime": 59.7694,
|
| 1147 |
+
"eval_samples_per_second": 695.774,
|
| 1148 |
+
"eval_steps_per_second": 0.686,
|
| 1149 |
+
"step": 3800
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 1.7853002550428936,
|
| 1153 |
+
"grad_norm": 0.05675831064581871,
|
| 1154 |
+
"learning_rate": 6.858312278301637e-05,
|
| 1155 |
+
"loss": 0.1303,
|
| 1156 |
+
"step": 3850
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 1.7853002550428936,
|
| 1160 |
+
"eval_loss": 0.1630547638293529,
|
| 1161 |
+
"eval_runtime": 59.779,
|
| 1162 |
+
"eval_samples_per_second": 695.662,
|
| 1163 |
+
"eval_steps_per_second": 0.686,
|
| 1164 |
+
"step": 3850
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.8084859726408533,
|
| 1168 |
+
"grad_norm": 0.04727062210440636,
|
| 1169 |
+
"learning_rate": 6.784431107959359e-05,
|
| 1170 |
+
"loss": 0.1312,
|
| 1171 |
+
"step": 3900
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.8084859726408533,
|
| 1175 |
+
"eval_loss": 0.1616409071893626,
|
| 1176 |
+
"eval_runtime": 59.6005,
|
| 1177 |
+
"eval_samples_per_second": 697.746,
|
| 1178 |
+
"eval_steps_per_second": 0.688,
|
| 1179 |
+
"step": 3900
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.8316716902388128,
|
| 1183 |
+
"grad_norm": 0.06378892064094543,
|
| 1184 |
+
"learning_rate": 6.710100716628344e-05,
|
| 1185 |
+
"loss": 0.1303,
|
| 1186 |
+
"step": 3950
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.8316716902388128,
|
| 1190 |
+
"eval_loss": 0.1622395658739077,
|
| 1191 |
+
"eval_runtime": 60.1499,
|
| 1192 |
+
"eval_samples_per_second": 691.373,
|
| 1193 |
+
"eval_steps_per_second": 0.682,
|
| 1194 |
+
"step": 3950
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 1.8548574078367726,
|
| 1198 |
+
"grad_norm": 0.05470576509833336,
|
| 1199 |
+
"learning_rate": 6.635339816587109e-05,
|
| 1200 |
+
"loss": 0.1308,
|
| 1201 |
+
"step": 4000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.8548574078367726,
|
| 1205 |
+
"eval_loss": 0.16317236170181762,
|
| 1206 |
+
"eval_runtime": 60.014,
|
| 1207 |
+
"eval_samples_per_second": 692.939,
|
| 1208 |
+
"eval_steps_per_second": 0.683,
|
| 1209 |
+
"step": 4000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 1.8780431254347323,
|
| 1213 |
+
"grad_norm": 0.053886763751506805,
|
| 1214 |
+
"learning_rate": 6.560167228492436e-05,
|
| 1215 |
+
"loss": 0.1297,
|
| 1216 |
+
"step": 4050
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 1.8780431254347323,
|
| 1220 |
+
"eval_loss": 0.16198886262197804,
|
| 1221 |
+
"eval_runtime": 60.8262,
|
| 1222 |
+
"eval_samples_per_second": 683.685,
|
| 1223 |
+
"eval_steps_per_second": 0.674,
|
| 1224 |
+
"step": 4050
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.9012288430326918,
|
| 1228 |
+
"grad_norm": 0.054583676159381866,
|
| 1229 |
+
"learning_rate": 6.484601876641375e-05,
|
| 1230 |
+
"loss": 0.1301,
|
| 1231 |
+
"step": 4100
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 1.9012288430326918,
|
| 1235 |
+
"eval_loss": 0.1616550050294764,
|
| 1236 |
+
"eval_runtime": 59.7779,
|
| 1237 |
+
"eval_samples_per_second": 695.675,
|
| 1238 |
+
"eval_steps_per_second": 0.686,
|
| 1239 |
+
"step": 4100
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 1.9244145606306515,
|
| 1243 |
+
"grad_norm": 0.071171335875988,
|
| 1244 |
+
"learning_rate": 6.408662784207149e-05,
|
| 1245 |
+
"loss": 0.131,
|
| 1246 |
+
"step": 4150
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 1.9244145606306515,
|
| 1250 |
+
"eval_loss": 0.15968682813566223,
|
| 1251 |
+
"eval_runtime": 60.227,
|
| 1252 |
+
"eval_samples_per_second": 690.487,
|
| 1253 |
+
"eval_steps_per_second": 0.681,
|
| 1254 |
+
"step": 4150
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 1.9476002782286113,
|
| 1258 |
+
"grad_norm": 0.05775531381368637,
|
| 1259 |
+
"learning_rate": 6.332369068450174e-05,
|
| 1260 |
+
"loss": 0.1296,
|
| 1261 |
+
"step": 4200
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 1.9476002782286113,
|
| 1265 |
+
"eval_loss": 0.16262199212265846,
|
| 1266 |
+
"eval_runtime": 60.3405,
|
| 1267 |
+
"eval_samples_per_second": 689.189,
|
| 1268 |
+
"eval_steps_per_second": 0.679,
|
| 1269 |
+
"step": 4200
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.9707859958265708,
|
| 1273 |
+
"grad_norm": 0.06425776332616806,
|
| 1274 |
+
"learning_rate": 6.255739935905396e-05,
|
| 1275 |
+
"loss": 0.1299,
|
| 1276 |
+
"step": 4250
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.9707859958265708,
|
| 1280 |
+
"eval_loss": 0.16324524366491053,
|
| 1281 |
+
"eval_runtime": 61.417,
|
| 1282 |
+
"eval_samples_per_second": 677.109,
|
| 1283 |
+
"eval_steps_per_second": 0.668,
|
| 1284 |
+
"step": 4250
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.9939717134245305,
|
| 1288 |
+
"grad_norm": 0.045762140303850174,
|
| 1289 |
+
"learning_rate": 6.178794677547137e-05,
|
| 1290 |
+
"loss": 0.1299,
|
| 1291 |
+
"step": 4300
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.9939717134245305,
|
| 1295 |
+
"eval_loss": 0.16053301797614244,
|
| 1296 |
+
"eval_runtime": 61.0801,
|
| 1297 |
+
"eval_samples_per_second": 680.844,
|
| 1298 |
+
"eval_steps_per_second": 0.671,
|
| 1299 |
+
"step": 4300
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.0171574310224902,
|
| 1303 |
+
"grad_norm": 0.07060451060533524,
|
| 1304 |
+
"learning_rate": 6.1015526639327035e-05,
|
| 1305 |
+
"loss": 0.1296,
|
| 1306 |
+
"step": 4350
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.0171574310224902,
|
| 1310 |
+
"eval_loss": 0.1620254674138633,
|
| 1311 |
+
"eval_runtime": 61.0829,
|
| 1312 |
+
"eval_samples_per_second": 680.812,
|
| 1313 |
+
"eval_steps_per_second": 0.671,
|
| 1314 |
+
"step": 4350
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.0403431486204497,
|
| 1318 |
+
"grad_norm": 0.059919316321611404,
|
| 1319 |
+
"learning_rate": 6.024033340325954e-05,
|
| 1320 |
+
"loss": 0.1302,
|
| 1321 |
+
"step": 4400
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 2.0403431486204497,
|
| 1325 |
+
"eval_loss": 0.16284223807997533,
|
| 1326 |
+
"eval_runtime": 61.5789,
|
| 1327 |
+
"eval_samples_per_second": 675.328,
|
| 1328 |
+
"eval_steps_per_second": 0.666,
|
| 1329 |
+
"step": 4400
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.0635288662184093,
|
| 1333 |
+
"grad_norm": 0.07983385026454926,
|
| 1334 |
+
"learning_rate": 5.946256221802051e-05,
|
| 1335 |
+
"loss": 0.13,
|
| 1336 |
+
"step": 4450
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 2.0635288662184093,
|
| 1340 |
+
"eval_loss": 0.16209282393788932,
|
| 1341 |
+
"eval_runtime": 61.6584,
|
| 1342 |
+
"eval_samples_per_second": 674.458,
|
| 1343 |
+
"eval_steps_per_second": 0.665,
|
| 1344 |
+
"step": 4450
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.086714583816369,
|
| 1348 |
+
"grad_norm": 0.07582173496484756,
|
| 1349 |
+
"learning_rate": 5.868240888334653e-05,
|
| 1350 |
+
"loss": 0.1296,
|
| 1351 |
+
"step": 4500
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 2.086714583816369,
|
| 1355 |
+
"eval_loss": 0.16158196377974565,
|
| 1356 |
+
"eval_runtime": 61.1826,
|
| 1357 |
+
"eval_samples_per_second": 679.703,
|
| 1358 |
+
"eval_steps_per_second": 0.67,
|
| 1359 |
+
"step": 4500
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.1099003014143287,
|
| 1363 |
+
"grad_norm": 0.06049995869398117,
|
| 1364 |
+
"learning_rate": 5.79000697986675e-05,
|
| 1365 |
+
"loss": 0.1298,
|
| 1366 |
+
"step": 4550
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 2.1099003014143287,
|
| 1370 |
+
"eval_loss": 0.16130609279963956,
|
| 1371 |
+
"eval_runtime": 61.0153,
|
| 1372 |
+
"eval_samples_per_second": 681.567,
|
| 1373 |
+
"eval_steps_per_second": 0.672,
|
| 1374 |
+
"step": 4550
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.1330860190122882,
|
| 1378 |
+
"grad_norm": 0.0440148264169693,
|
| 1379 |
+
"learning_rate": 5.7115741913664264e-05,
|
| 1380 |
+
"loss": 0.1299,
|
| 1381 |
+
"step": 4600
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 2.1330860190122882,
|
| 1385 |
+
"eval_loss": 0.16027799763638953,
|
| 1386 |
+
"eval_runtime": 61.1993,
|
| 1387 |
+
"eval_samples_per_second": 679.517,
|
| 1388 |
+
"eval_steps_per_second": 0.67,
|
| 1389 |
+
"step": 4600
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.156271736610248,
|
| 1393 |
+
"grad_norm": 0.05254065990447998,
|
| 1394 |
+
"learning_rate": 5.6329622678687463e-05,
|
| 1395 |
+
"loss": 0.1299,
|
| 1396 |
+
"step": 4650
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.156271736610248,
|
| 1400 |
+
"eval_loss": 0.16206274484291652,
|
| 1401 |
+
"eval_runtime": 61.4415,
|
| 1402 |
+
"eval_samples_per_second": 676.839,
|
| 1403 |
+
"eval_steps_per_second": 0.667,
|
| 1404 |
+
"step": 4650
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.1794574542082077,
|
| 1408 |
+
"grad_norm": 0.06294432282447815,
|
| 1409 |
+
"learning_rate": 5.5541909995050554e-05,
|
| 1410 |
+
"loss": 0.1306,
|
| 1411 |
+
"step": 4700
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.1794574542082077,
|
| 1415 |
+
"eval_loss": 0.16140170723024802,
|
| 1416 |
+
"eval_runtime": 60.8861,
|
| 1417 |
+
"eval_samples_per_second": 683.013,
|
| 1418 |
+
"eval_steps_per_second": 0.673,
|
| 1419 |
+
"step": 4700
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.202643171806167,
|
| 1423 |
+
"grad_norm": 0.06710942089557648,
|
| 1424 |
+
"learning_rate": 5.475280216520913e-05,
|
| 1425 |
+
"loss": 0.1303,
|
| 1426 |
+
"step": 4750
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 2.202643171806167,
|
| 1430 |
+
"eval_loss": 0.16245448075670843,
|
| 1431 |
+
"eval_runtime": 61.2839,
|
| 1432 |
+
"eval_samples_per_second": 678.58,
|
| 1433 |
+
"eval_steps_per_second": 0.669,
|
| 1434 |
+
"step": 4750
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.225828889404127,
|
| 1438 |
+
"grad_norm": 0.05298132076859474,
|
| 1439 |
+
"learning_rate": 5.396249784283942e-05,
|
| 1440 |
+
"loss": 0.13,
|
| 1441 |
+
"step": 4800
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 2.225828889404127,
|
| 1445 |
+
"eval_loss": 0.1623738898660767,
|
| 1446 |
+
"eval_runtime": 61.1531,
|
| 1447 |
+
"eval_samples_per_second": 680.031,
|
| 1448 |
+
"eval_steps_per_second": 0.67,
|
| 1449 |
+
"step": 4800
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.2490146070020867,
|
| 1453 |
+
"grad_norm": 0.04066763445734978,
|
| 1454 |
+
"learning_rate": 5.317119598282823e-05,
|
| 1455 |
+
"loss": 0.1295,
|
| 1456 |
+
"step": 4850
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 2.2490146070020867,
|
| 1460 |
+
"eval_loss": 0.1627438727811327,
|
| 1461 |
+
"eval_runtime": 61.0414,
|
| 1462 |
+
"eval_samples_per_second": 681.275,
|
| 1463 |
+
"eval_steps_per_second": 0.672,
|
| 1464 |
+
"step": 4850
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 2.272200324600046,
|
| 1468 |
+
"grad_norm": 0.061821240931749344,
|
| 1469 |
+
"learning_rate": 5.2379095791187124e-05,
|
| 1470 |
+
"loss": 0.1299,
|
| 1471 |
+
"step": 4900
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 2.272200324600046,
|
| 1475 |
+
"eval_loss": 0.16086717177928397,
|
| 1476 |
+
"eval_runtime": 60.7945,
|
| 1477 |
+
"eval_samples_per_second": 684.042,
|
| 1478 |
+
"eval_steps_per_second": 0.674,
|
| 1479 |
+
"step": 4900
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 2.295386042198006,
|
| 1483 |
+
"grad_norm": 0.08038394153118134,
|
| 1484 |
+
"learning_rate": 5.158639667490339e-05,
|
| 1485 |
+
"loss": 0.13,
|
| 1486 |
+
"step": 4950
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.295386042198006,
|
| 1490 |
+
"eval_loss": 0.16221664317086187,
|
| 1491 |
+
"eval_runtime": 61.6742,
|
| 1492 |
+
"eval_samples_per_second": 674.285,
|
| 1493 |
+
"eval_steps_per_second": 0.665,
|
| 1494 |
+
"step": 4950
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.3185717597959656,
|
| 1498 |
+
"grad_norm": 0.0556926503777504,
|
| 1499 |
+
"learning_rate": 5.0793298191740404e-05,
|
| 1500 |
+
"loss": 0.1311,
|
| 1501 |
+
"step": 5000
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 2.3185717597959656,
|
| 1505 |
+
"eval_loss": 0.16015339844546791,
|
| 1506 |
+
"eval_runtime": 61.3657,
|
| 1507 |
+
"eval_samples_per_second": 677.675,
|
| 1508 |
+
"eval_steps_per_second": 0.668,
|
| 1509 |
+
"step": 5000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 2.3417574773939256,
|
| 1513 |
+
"grad_norm": 0.06645477563142776,
|
| 1514 |
+
"learning_rate": 5e-05,
|
| 1515 |
+
"loss": 0.1284,
|
| 1516 |
+
"step": 5050
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.3417574773939256,
|
| 1520 |
+
"eval_loss": 0.16160674186313023,
|
| 1521 |
+
"eval_runtime": 61.4737,
|
| 1522 |
+
"eval_samples_per_second": 676.484,
|
| 1523 |
+
"eval_steps_per_second": 0.667,
|
| 1524 |
+
"step": 5050
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 2.364943194991885,
|
| 1528 |
+
"grad_norm": 0.05365500971674919,
|
| 1529 |
+
"learning_rate": 4.92067018082596e-05,
|
| 1530 |
+
"loss": 0.13,
|
| 1531 |
+
"step": 5100
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 2.364943194991885,
|
| 1535 |
+
"eval_loss": 0.16016484459556096,
|
| 1536 |
+
"eval_runtime": 61.4058,
|
| 1537 |
+
"eval_samples_per_second": 677.232,
|
| 1538 |
+
"eval_steps_per_second": 0.668,
|
| 1539 |
+
"step": 5100
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 2.3881289125898446,
|
| 1543 |
+
"grad_norm": 0.0499204620718956,
|
| 1544 |
+
"learning_rate": 4.841360332509663e-05,
|
| 1545 |
+
"loss": 0.129,
|
| 1546 |
+
"step": 5150
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 2.3881289125898446,
|
| 1550 |
+
"eval_loss": 0.16054727464378063,
|
| 1551 |
+
"eval_runtime": 61.1539,
|
| 1552 |
+
"eval_samples_per_second": 680.023,
|
| 1553 |
+
"eval_steps_per_second": 0.67,
|
| 1554 |
+
"step": 5150
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 2.411314630187804,
|
| 1558 |
+
"grad_norm": 0.07284457236528397,
|
| 1559 |
+
"learning_rate": 4.762090420881289e-05,
|
| 1560 |
+
"loss": 0.129,
|
| 1561 |
+
"step": 5200
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 2.411314630187804,
|
| 1565 |
+
"eval_loss": 0.16057287778830004,
|
| 1566 |
+
"eval_runtime": 60.5785,
|
| 1567 |
+
"eval_samples_per_second": 686.481,
|
| 1568 |
+
"eval_steps_per_second": 0.677,
|
| 1569 |
+
"step": 5200
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 2.434500347785764,
|
| 1573 |
+
"grad_norm": 0.06511891633272171,
|
| 1574 |
+
"learning_rate": 4.6828804017171776e-05,
|
| 1575 |
+
"loss": 0.1297,
|
| 1576 |
+
"step": 5250
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 2.434500347785764,
|
| 1580 |
+
"eval_loss": 0.16202011190836896,
|
| 1581 |
+
"eval_runtime": 61.4053,
|
| 1582 |
+
"eval_samples_per_second": 677.238,
|
| 1583 |
+
"eval_steps_per_second": 0.668,
|
| 1584 |
+
"step": 5250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.4576860653837236,
|
| 1588 |
+
"grad_norm": 0.05936937406659126,
|
| 1589 |
+
"learning_rate": 4.603750215716057e-05,
|
| 1590 |
+
"loss": 0.1293,
|
| 1591 |
+
"step": 5300
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 2.4576860653837236,
|
| 1595 |
+
"eval_loss": 0.16067086041480225,
|
| 1596 |
+
"eval_runtime": 60.4469,
|
| 1597 |
+
"eval_samples_per_second": 687.976,
|
| 1598 |
+
"eval_steps_per_second": 0.678,
|
| 1599 |
+
"step": 5300
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 2.480871782981683,
|
| 1603 |
+
"grad_norm": 0.039836496114730835,
|
| 1604 |
+
"learning_rate": 4.5247197834790876e-05,
|
| 1605 |
+
"loss": 0.1288,
|
| 1606 |
+
"step": 5350
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 2.480871782981683,
|
| 1610 |
+
"eval_loss": 0.1614640227625451,
|
| 1611 |
+
"eval_runtime": 60.9513,
|
| 1612 |
+
"eval_samples_per_second": 682.283,
|
| 1613 |
+
"eval_steps_per_second": 0.673,
|
| 1614 |
+
"step": 5350
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 2.504057500579643,
|
| 1618 |
+
"grad_norm": 0.04305760934948921,
|
| 1619 |
+
"learning_rate": 4.445809000494946e-05,
|
| 1620 |
+
"loss": 0.1294,
|
| 1621 |
+
"step": 5400
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 2.504057500579643,
|
| 1625 |
+
"eval_loss": 0.16139181990447046,
|
| 1626 |
+
"eval_runtime": 60.6766,
|
| 1627 |
+
"eval_samples_per_second": 685.371,
|
| 1628 |
+
"eval_steps_per_second": 0.676,
|
| 1629 |
+
"step": 5400
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 2.5272432181776026,
|
| 1633 |
+
"grad_norm": 0.06780368089675903,
|
| 1634 |
+
"learning_rate": 4.3670377321312535e-05,
|
| 1635 |
+
"loss": 0.1285,
|
| 1636 |
+
"step": 5450
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 2.5272432181776026,
|
| 1640 |
+
"eval_loss": 0.1619736397134425,
|
| 1641 |
+
"eval_runtime": 60.7281,
|
| 1642 |
+
"eval_samples_per_second": 684.79,
|
| 1643 |
+
"eval_steps_per_second": 0.675,
|
| 1644 |
+
"step": 5450
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 2.550428935775562,
|
| 1648 |
+
"grad_norm": 0.052273835986852646,
|
| 1649 |
+
"learning_rate": 4.288425808633575e-05,
|
| 1650 |
+
"loss": 0.1303,
|
| 1651 |
+
"step": 5500
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 2.550428935775562,
|
| 1655 |
+
"eval_loss": 0.16178818674979198,
|
| 1656 |
+
"eval_runtime": 60.8875,
|
| 1657 |
+
"eval_samples_per_second": 682.997,
|
| 1658 |
+
"eval_steps_per_second": 0.673,
|
| 1659 |
+
"step": 5500
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 2.573614653373522,
|
| 1663 |
+
"grad_norm": 0.045574627816677094,
|
| 1664 |
+
"learning_rate": 4.20999302013325e-05,
|
| 1665 |
+
"loss": 0.1291,
|
| 1666 |
+
"step": 5550
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 2.573614653373522,
|
| 1670 |
+
"eval_loss": 0.16034006952877458,
|
| 1671 |
+
"eval_runtime": 60.7378,
|
| 1672 |
+
"eval_samples_per_second": 684.681,
|
| 1673 |
+
"eval_steps_per_second": 0.675,
|
| 1674 |
+
"step": 5550
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 2.5968003709714815,
|
| 1678 |
+
"grad_norm": 0.044092051684856415,
|
| 1679 |
+
"learning_rate": 4.131759111665349e-05,
|
| 1680 |
+
"loss": 0.1298,
|
| 1681 |
+
"step": 5600
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 2.5968003709714815,
|
| 1685 |
+
"eval_loss": 0.16090484909780667,
|
| 1686 |
+
"eval_runtime": 60.4675,
|
| 1687 |
+
"eval_samples_per_second": 687.741,
|
| 1688 |
+
"eval_steps_per_second": 0.678,
|
| 1689 |
+
"step": 5600
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 2.6199860885694415,
|
| 1693 |
+
"grad_norm": 0.05473971739411354,
|
| 1694 |
+
"learning_rate": 4.0537437781979506e-05,
|
| 1695 |
+
"loss": 0.1288,
|
| 1696 |
+
"step": 5650
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 2.6199860885694415,
|
| 1700 |
+
"eval_loss": 0.1604315377337276,
|
| 1701 |
+
"eval_runtime": 62.8239,
|
| 1702 |
+
"eval_samples_per_second": 661.946,
|
| 1703 |
+
"eval_steps_per_second": 0.653,
|
| 1704 |
+
"step": 5650
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.643171806167401,
|
| 1708 |
+
"grad_norm": 0.07100555300712585,
|
| 1709 |
+
"learning_rate": 3.9759666596740476e-05,
|
| 1710 |
+
"loss": 0.129,
|
| 1711 |
+
"step": 5700
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.643171806167401,
|
| 1715 |
+
"eval_loss": 0.15997494100305837,
|
| 1716 |
+
"eval_runtime": 61.3008,
|
| 1717 |
+
"eval_samples_per_second": 678.392,
|
| 1718 |
+
"eval_steps_per_second": 0.669,
|
| 1719 |
+
"step": 5700
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 2.6663575237653605,
|
| 1723 |
+
"grad_norm": 0.04020215570926666,
|
| 1724 |
+
"learning_rate": 3.898447336067297e-05,
|
| 1725 |
+
"loss": 0.1291,
|
| 1726 |
+
"step": 5750
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 2.6663575237653605,
|
| 1730 |
+
"eval_loss": 0.1596748490832133,
|
| 1731 |
+
"eval_runtime": 60.6148,
|
| 1732 |
+
"eval_samples_per_second": 686.07,
|
| 1733 |
+
"eval_steps_per_second": 0.676,
|
| 1734 |
+
"step": 5750
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 2.68954324136332,
|
| 1738 |
+
"grad_norm": 0.05526584014296532,
|
| 1739 |
+
"learning_rate": 3.821205322452863e-05,
|
| 1740 |
+
"loss": 0.1291,
|
| 1741 |
+
"step": 5800
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 2.68954324136332,
|
| 1745 |
+
"eval_loss": 0.16091962633426782,
|
| 1746 |
+
"eval_runtime": 60.1717,
|
| 1747 |
+
"eval_samples_per_second": 691.122,
|
| 1748 |
+
"eval_steps_per_second": 0.681,
|
| 1749 |
+
"step": 5800
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 2.71272895896128,
|
| 1753 |
+
"grad_norm": 0.052167922258377075,
|
| 1754 |
+
"learning_rate": 3.744260064094604e-05,
|
| 1755 |
+
"loss": 0.129,
|
| 1756 |
+
"step": 5850
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 2.71272895896128,
|
| 1760 |
+
"eval_loss": 0.16112806362615253,
|
| 1761 |
+
"eval_runtime": 60.1273,
|
| 1762 |
+
"eval_samples_per_second": 691.633,
|
| 1763 |
+
"eval_steps_per_second": 0.682,
|
| 1764 |
+
"step": 5850
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 2.7359146765592395,
|
| 1768 |
+
"grad_norm": 0.054320793598890305,
|
| 1769 |
+
"learning_rate": 3.6676309315498256e-05,
|
| 1770 |
+
"loss": 0.13,
|
| 1771 |
+
"step": 5900
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 2.7359146765592395,
|
| 1775 |
+
"eval_loss": 0.15996250695505343,
|
| 1776 |
+
"eval_runtime": 60.655,
|
| 1777 |
+
"eval_samples_per_second": 685.616,
|
| 1778 |
+
"eval_steps_per_second": 0.676,
|
| 1779 |
+
"step": 5900
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 2.7591003941571994,
|
| 1783 |
+
"grad_norm": 0.05470626428723335,
|
| 1784 |
+
"learning_rate": 3.591337215792852e-05,
|
| 1785 |
+
"loss": 0.1296,
|
| 1786 |
+
"step": 5950
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 2.7591003941571994,
|
| 1790 |
+
"eval_loss": 0.16025288890609335,
|
| 1791 |
+
"eval_runtime": 60.826,
|
| 1792 |
+
"eval_samples_per_second": 683.688,
|
| 1793 |
+
"eval_steps_per_second": 0.674,
|
| 1794 |
+
"step": 5950
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 2.782286111755159,
|
| 1798 |
+
"grad_norm": 0.04805810749530792,
|
| 1799 |
+
"learning_rate": 3.515398123358627e-05,
|
| 1800 |
+
"loss": 0.1294,
|
| 1801 |
+
"step": 6000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 2.782286111755159,
|
| 1805 |
+
"eval_loss": 0.15918263724182835,
|
| 1806 |
+
"eval_runtime": 60.2321,
|
| 1807 |
+
"eval_samples_per_second": 690.429,
|
| 1808 |
+
"eval_steps_per_second": 0.681,
|
| 1809 |
+
"step": 6000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.8054718293531185,
|
| 1813 |
+
"grad_norm": 0.04185302183032036,
|
| 1814 |
+
"learning_rate": 3.439832771507565e-05,
|
| 1815 |
+
"loss": 0.1283,
|
| 1816 |
+
"step": 6050
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.8054718293531185,
|
| 1820 |
+
"eval_loss": 0.16179385240233157,
|
| 1821 |
+
"eval_runtime": 60.9176,
|
| 1822 |
+
"eval_samples_per_second": 682.66,
|
| 1823 |
+
"eval_steps_per_second": 0.673,
|
| 1824 |
+
"step": 6050
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 2.828657546951078,
|
| 1828 |
+
"grad_norm": 0.04609336704015732,
|
| 1829 |
+
"learning_rate": 3.364660183412892e-05,
|
| 1830 |
+
"loss": 0.1292,
|
| 1831 |
+
"step": 6100
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 2.828657546951078,
|
| 1835 |
+
"eval_loss": 0.1611929898635588,
|
| 1836 |
+
"eval_runtime": 60.5916,
|
| 1837 |
+
"eval_samples_per_second": 686.333,
|
| 1838 |
+
"eval_steps_per_second": 0.677,
|
| 1839 |
+
"step": 6100
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 2.851843264549038,
|
| 1843 |
+
"grad_norm": 0.05404876172542572,
|
| 1844 |
+
"learning_rate": 3.289899283371657e-05,
|
| 1845 |
+
"loss": 0.128,
|
| 1846 |
+
"step": 6150
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 2.851843264549038,
|
| 1850 |
+
"eval_loss": 0.16039360794951976,
|
| 1851 |
+
"eval_runtime": 60.5961,
|
| 1852 |
+
"eval_samples_per_second": 686.282,
|
| 1853 |
+
"eval_steps_per_second": 0.677,
|
| 1854 |
+
"step": 6150
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 2.8750289821469974,
|
| 1858 |
+
"grad_norm": 0.06787659227848053,
|
| 1859 |
+
"learning_rate": 3.215568892040641e-05,
|
| 1860 |
+
"loss": 0.1288,
|
| 1861 |
+
"step": 6200
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 2.8750289821469974,
|
| 1865 |
+
"eval_loss": 0.16113480515361805,
|
| 1866 |
+
"eval_runtime": 60.2775,
|
| 1867 |
+
"eval_samples_per_second": 689.909,
|
| 1868 |
+
"eval_steps_per_second": 0.68,
|
| 1869 |
+
"step": 6200
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 2.8982146997449574,
|
| 1873 |
+
"grad_norm": 0.06937435269355774,
|
| 1874 |
+
"learning_rate": 3.141687721698363e-05,
|
| 1875 |
+
"loss": 0.1283,
|
| 1876 |
+
"step": 6250
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 2.8982146997449574,
|
| 1880 |
+
"eval_loss": 0.16087572214972407,
|
| 1881 |
+
"eval_runtime": 60.6789,
|
| 1882 |
+
"eval_samples_per_second": 685.345,
|
| 1883 |
+
"eval_steps_per_second": 0.676,
|
| 1884 |
+
"step": 6250
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 2.921400417342917,
|
| 1888 |
+
"grad_norm": 0.08074232190847397,
|
| 1889 |
+
"learning_rate": 3.0682743715343564e-05,
|
| 1890 |
+
"loss": 0.1292,
|
| 1891 |
+
"step": 6300
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 2.921400417342917,
|
| 1895 |
+
"eval_loss": 0.16049740787316144,
|
| 1896 |
+
"eval_runtime": 60.3194,
|
| 1897 |
+
"eval_samples_per_second": 689.43,
|
| 1898 |
+
"eval_steps_per_second": 0.68,
|
| 1899 |
+
"step": 6300
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 2.9445861349408764,
|
| 1903 |
+
"grad_norm": 0.03976515680551529,
|
| 1904 |
+
"learning_rate": 2.9953473229669328e-05,
|
| 1905 |
+
"loss": 0.1302,
|
| 1906 |
+
"step": 6350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.9445861349408764,
|
| 1910 |
+
"eval_loss": 0.16023700059761273,
|
| 1911 |
+
"eval_runtime": 60.8537,
|
| 1912 |
+
"eval_samples_per_second": 683.377,
|
| 1913 |
+
"eval_steps_per_second": 0.674,
|
| 1914 |
+
"step": 6350
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.967771852538836,
|
| 1918 |
+
"grad_norm": 0.05303976684808731,
|
| 1919 |
+
"learning_rate": 2.9229249349905684e-05,
|
| 1920 |
+
"loss": 0.1285,
|
| 1921 |
+
"step": 6400
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.967771852538836,
|
| 1925 |
+
"eval_loss": 0.1601465398516622,
|
| 1926 |
+
"eval_runtime": 60.6472,
|
| 1927 |
+
"eval_samples_per_second": 685.703,
|
| 1928 |
+
"eval_steps_per_second": 0.676,
|
| 1929 |
+
"step": 6400
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 2.990957570136796,
|
| 1933 |
+
"grad_norm": 0.0519745759665966,
|
| 1934 |
+
"learning_rate": 2.851025439554142e-05,
|
| 1935 |
+
"loss": 0.1286,
|
| 1936 |
+
"step": 6450
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 2.990957570136796,
|
| 1940 |
+
"eval_loss": 0.16085429229133483,
|
| 1941 |
+
"eval_runtime": 60.2507,
|
| 1942 |
+
"eval_samples_per_second": 690.216,
|
| 1943 |
+
"eval_steps_per_second": 0.68,
|
| 1944 |
+
"step": 6450
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 3.0141432877347554,
|
| 1948 |
+
"grad_norm": 0.050518251955509186,
|
| 1949 |
+
"learning_rate": 2.7796669369711294e-05,
|
| 1950 |
+
"loss": 0.1301,
|
| 1951 |
+
"step": 6500
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 3.0141432877347554,
|
| 1955 |
+
"eval_loss": 0.16015394660421692,
|
| 1956 |
+
"eval_runtime": 60.5015,
|
| 1957 |
+
"eval_samples_per_second": 687.355,
|
| 1958 |
+
"eval_steps_per_second": 0.678,
|
| 1959 |
+
"step": 6500
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 3.037329005332715,
|
| 1963 |
+
"grad_norm": 0.04253960773348808,
|
| 1964 |
+
"learning_rate": 2.708867391362948e-05,
|
| 1965 |
+
"loss": 0.1296,
|
| 1966 |
+
"step": 6550
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 3.037329005332715,
|
| 1970 |
+
"eval_loss": 0.1597283595131218,
|
| 1971 |
+
"eval_runtime": 60.13,
|
| 1972 |
+
"eval_samples_per_second": 691.601,
|
| 1973 |
+
"eval_steps_per_second": 0.682,
|
| 1974 |
+
"step": 6550
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 3.060514722930675,
|
| 1978 |
+
"grad_norm": 0.06899340450763702,
|
| 1979 |
+
"learning_rate": 2.638644626136587e-05,
|
| 1980 |
+
"loss": 0.1291,
|
| 1981 |
+
"step": 6600
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 3.060514722930675,
|
| 1985 |
+
"eval_loss": 0.1604277250117246,
|
| 1986 |
+
"eval_runtime": 60.4618,
|
| 1987 |
+
"eval_samples_per_second": 687.806,
|
| 1988 |
+
"eval_steps_per_second": 0.678,
|
| 1989 |
+
"step": 6600
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 3.0837004405286343,
|
| 1993 |
+
"grad_norm": 0.06556117534637451,
|
| 1994 |
+
"learning_rate": 2.5690163194976575e-05,
|
| 1995 |
+
"loss": 0.1288,
|
| 1996 |
+
"step": 6650
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 3.0837004405286343,
|
| 2000 |
+
"eval_loss": 0.15953636330193482,
|
| 2001 |
+
"eval_runtime": 60.2757,
|
| 2002 |
+
"eval_samples_per_second": 689.93,
|
| 2003 |
+
"eval_steps_per_second": 0.68,
|
| 2004 |
+
"step": 6650
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 3.106886158126594,
|
| 2008 |
+
"grad_norm": 0.03685734421014786,
|
| 2009 |
+
"learning_rate": 2.500000000000001e-05,
|
| 2010 |
+
"loss": 0.129,
|
| 2011 |
+
"step": 6700
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 3.106886158126594,
|
| 2015 |
+
"eval_loss": 0.159308270335797,
|
| 2016 |
+
"eval_runtime": 60.624,
|
| 2017 |
+
"eval_samples_per_second": 685.966,
|
| 2018 |
+
"eval_steps_per_second": 0.676,
|
| 2019 |
+
"step": 6700
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 3.130071875724554,
|
| 2023 |
+
"grad_norm": 0.0451020672917366,
|
| 2024 |
+
"learning_rate": 2.4316130421329697e-05,
|
| 2025 |
+
"loss": 0.1286,
|
| 2026 |
+
"step": 6750
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 3.130071875724554,
|
| 2030 |
+
"eval_loss": 0.15995884031774596,
|
| 2031 |
+
"eval_runtime": 60.3654,
|
| 2032 |
+
"eval_samples_per_second": 688.905,
|
| 2033 |
+
"eval_steps_per_second": 0.679,
|
| 2034 |
+
"step": 6750
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 3.1532575933225133,
|
| 2038 |
+
"grad_norm": 0.0495733842253685,
|
| 2039 |
+
"learning_rate": 2.363872661947488e-05,
|
| 2040 |
+
"loss": 0.1293,
|
| 2041 |
+
"step": 6800
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 3.1532575933225133,
|
| 2045 |
+
"eval_loss": 0.15987331824692497,
|
| 2046 |
+
"eval_runtime": 60.4636,
|
| 2047 |
+
"eval_samples_per_second": 687.786,
|
| 2048 |
+
"eval_steps_per_second": 0.678,
|
| 2049 |
+
"step": 6800
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 3.176443310920473,
|
| 2053 |
+
"grad_norm": 0.05756652355194092,
|
| 2054 |
+
"learning_rate": 2.296795912722014e-05,
|
| 2055 |
+
"loss": 0.1289,
|
| 2056 |
+
"step": 6850
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 3.176443310920473,
|
| 2060 |
+
"eval_loss": 0.15986134614331013,
|
| 2061 |
+
"eval_runtime": 61.0063,
|
| 2062 |
+
"eval_samples_per_second": 681.667,
|
| 2063 |
+
"eval_steps_per_second": 0.672,
|
| 2064 |
+
"step": 6850
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 3.199629028518433,
|
| 2068 |
+
"grad_norm": 0.0467820018529892,
|
| 2069 |
+
"learning_rate": 2.2303996806694488e-05,
|
| 2070 |
+
"loss": 0.1295,
|
| 2071 |
+
"step": 6900
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 3.199629028518433,
|
| 2075 |
+
"eval_loss": 0.16011030076900337,
|
| 2076 |
+
"eval_runtime": 60.1041,
|
| 2077 |
+
"eval_samples_per_second": 691.9,
|
| 2078 |
+
"eval_steps_per_second": 0.682,
|
| 2079 |
+
"step": 6900
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 3.2228147461163923,
|
| 2083 |
+
"grad_norm": 0.04179982468485832,
|
| 2084 |
+
"learning_rate": 2.164700680686147e-05,
|
| 2085 |
+
"loss": 0.1287,
|
| 2086 |
+
"step": 6950
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 3.2228147461163923,
|
| 2090 |
+
"eval_loss": 0.15917751068552838,
|
| 2091 |
+
"eval_runtime": 60.5321,
|
| 2092 |
+
"eval_samples_per_second": 687.007,
|
| 2093 |
+
"eval_steps_per_second": 0.677,
|
| 2094 |
+
"step": 6950
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 3.246000463714352,
|
| 2098 |
+
"grad_norm": 0.053910572081804276,
|
| 2099 |
+
"learning_rate": 2.09971545214401e-05,
|
| 2100 |
+
"loss": 0.1286,
|
| 2101 |
+
"step": 7000
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 3.246000463714352,
|
| 2105 |
+
"eval_loss": 0.15998092838627764,
|
| 2106 |
+
"eval_runtime": 60.4067,
|
| 2107 |
+
"eval_samples_per_second": 688.434,
|
| 2108 |
+
"eval_steps_per_second": 0.679,
|
| 2109 |
+
"step": 7000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 3.2691861813123118,
|
| 2113 |
+
"grad_norm": 0.04404950886964798,
|
| 2114 |
+
"learning_rate": 2.0354603547267985e-05,
|
| 2115 |
+
"loss": 0.1283,
|
| 2116 |
+
"step": 7050
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 3.2691861813123118,
|
| 2120 |
+
"eval_loss": 0.1597617331551387,
|
| 2121 |
+
"eval_runtime": 60.4218,
|
| 2122 |
+
"eval_samples_per_second": 688.262,
|
| 2123 |
+
"eval_steps_per_second": 0.679,
|
| 2124 |
+
"step": 7050
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 3.2923718989102713,
|
| 2128 |
+
"grad_norm": 0.04763752967119217,
|
| 2129 |
+
"learning_rate": 1.9719515643116674e-05,
|
| 2130 |
+
"loss": 0.1288,
|
| 2131 |
+
"step": 7100
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 3.2923718989102713,
|
| 2135 |
+
"eval_loss": 0.16116006530852447,
|
| 2136 |
+
"eval_runtime": 60.2132,
|
| 2137 |
+
"eval_samples_per_second": 690.646,
|
| 2138 |
+
"eval_steps_per_second": 0.681,
|
| 2139 |
+
"step": 7100
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 3.3155576165082308,
|
| 2143 |
+
"grad_norm": 0.049567196518182755,
|
| 2144 |
+
"learning_rate": 1.9092050688969738e-05,
|
| 2145 |
+
"loss": 0.1298,
|
| 2146 |
+
"step": 7150
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 3.3155576165082308,
|
| 2150 |
+
"eval_loss": 0.15965543804361845,
|
| 2151 |
+
"eval_runtime": 60.3928,
|
| 2152 |
+
"eval_samples_per_second": 688.592,
|
| 2153 |
+
"eval_steps_per_second": 0.679,
|
| 2154 |
+
"step": 7150
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 3.3387433341061907,
|
| 2158 |
+
"grad_norm": 0.05488676205277443,
|
| 2159 |
+
"learning_rate": 1.847236664577389e-05,
|
| 2160 |
+
"loss": 0.1284,
|
| 2161 |
+
"step": 7200
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 3.3387433341061907,
|
| 2165 |
+
"eval_loss": 0.16050384662882064,
|
| 2166 |
+
"eval_runtime": 60.121,
|
| 2167 |
+
"eval_samples_per_second": 691.705,
|
| 2168 |
+
"eval_steps_per_second": 0.682,
|
| 2169 |
+
"step": 7200
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 3.3619290517041502,
|
| 2173 |
+
"grad_norm": 0.04124298691749573,
|
| 2174 |
+
"learning_rate": 1.7860619515673033e-05,
|
| 2175 |
+
"loss": 0.1289,
|
| 2176 |
+
"step": 7250
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 3.3619290517041502,
|
| 2180 |
+
"eval_loss": 0.16054145931691394,
|
| 2181 |
+
"eval_runtime": 60.2046,
|
| 2182 |
+
"eval_samples_per_second": 690.745,
|
| 2183 |
+
"eval_steps_per_second": 0.681,
|
| 2184 |
+
"step": 7250
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 3.3851147693021097,
|
| 2188 |
+
"grad_norm": 0.04400424286723137,
|
| 2189 |
+
"learning_rate": 1.725696330273575e-05,
|
| 2190 |
+
"loss": 0.1289,
|
| 2191 |
+
"step": 7300
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 3.3851147693021097,
|
| 2195 |
+
"eval_loss": 0.15999099129576416,
|
| 2196 |
+
"eval_runtime": 60.4869,
|
| 2197 |
+
"eval_samples_per_second": 687.52,
|
| 2198 |
+
"eval_steps_per_second": 0.678,
|
| 2199 |
+
"step": 7300
|
| 2200 |
+
},
|
| 2201 |
+
{
|
| 2202 |
+
"epoch": 3.4083004869000697,
|
| 2203 |
+
"grad_norm": 0.05488509312272072,
|
| 2204 |
+
"learning_rate": 1.6661549974185424e-05,
|
| 2205 |
+
"loss": 0.1285,
|
| 2206 |
+
"step": 7350
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 3.4083004869000697,
|
| 2210 |
+
"eval_loss": 0.16051823730892306,
|
| 2211 |
+
"eval_runtime": 60.1981,
|
| 2212 |
+
"eval_samples_per_second": 690.819,
|
| 2213 |
+
"eval_steps_per_second": 0.681,
|
| 2214 |
+
"step": 7350
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 3.431486204498029,
|
| 2218 |
+
"grad_norm": 0.06722457706928253,
|
| 2219 |
+
"learning_rate": 1.60745294221434e-05,
|
| 2220 |
+
"loss": 0.1286,
|
| 2221 |
+
"step": 7400
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 3.431486204498029,
|
| 2225 |
+
"eval_loss": 0.1610307768591294,
|
| 2226 |
+
"eval_runtime": 60.7755,
|
| 2227 |
+
"eval_samples_per_second": 684.256,
|
| 2228 |
+
"eval_steps_per_second": 0.675,
|
| 2229 |
+
"step": 7400
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 3.4546719220959887,
|
| 2233 |
+
"grad_norm": 0.04814394935965538,
|
| 2234 |
+
"learning_rate": 1.549604942589441e-05,
|
| 2235 |
+
"loss": 0.1278,
|
| 2236 |
+
"step": 7450
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 3.4546719220959887,
|
| 2240 |
+
"eval_loss": 0.1598065741965525,
|
| 2241 |
+
"eval_runtime": 59.9968,
|
| 2242 |
+
"eval_samples_per_second": 693.136,
|
| 2243 |
+
"eval_steps_per_second": 0.683,
|
| 2244 |
+
"step": 7450
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 3.4778576396939487,
|
| 2248 |
+
"grad_norm": 0.04934167116880417,
|
| 2249 |
+
"learning_rate": 1.4926255614683932e-05,
|
| 2250 |
+
"loss": 0.1274,
|
| 2251 |
+
"step": 7500
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 3.4778576396939487,
|
| 2255 |
+
"eval_loss": 0.15982454893723042,
|
| 2256 |
+
"eval_runtime": 60.201,
|
| 2257 |
+
"eval_samples_per_second": 690.786,
|
| 2258 |
+
"eval_steps_per_second": 0.681,
|
| 2259 |
+
"step": 7500
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 3.501043357291908,
|
| 2263 |
+
"grad_norm": 0.04529615864157677,
|
| 2264 |
+
"learning_rate": 1.4365291431056871e-05,
|
| 2265 |
+
"loss": 0.1297,
|
| 2266 |
+
"step": 7550
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 3.501043357291908,
|
| 2270 |
+
"eval_loss": 0.15986133524024926,
|
| 2271 |
+
"eval_runtime": 59.95,
|
| 2272 |
+
"eval_samples_per_second": 693.678,
|
| 2273 |
+
"eval_steps_per_second": 0.684,
|
| 2274 |
+
"step": 7550
|
| 2275 |
+
},
|
| 2276 |
+
{
|
| 2277 |
+
"epoch": 3.5242290748898677,
|
| 2278 |
+
"grad_norm": 0.0399620421230793,
|
| 2279 |
+
"learning_rate": 1.3813298094746491e-05,
|
| 2280 |
+
"loss": 0.1288,
|
| 2281 |
+
"step": 7600
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 3.5242290748898677,
|
| 2285 |
+
"eval_loss": 0.15905609221590689,
|
| 2286 |
+
"eval_runtime": 61.181,
|
| 2287 |
+
"eval_samples_per_second": 679.72,
|
| 2288 |
+
"eval_steps_per_second": 0.67,
|
| 2289 |
+
"step": 7600
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 3.5474147924878277,
|
| 2293 |
+
"grad_norm": 0.05973295867443085,
|
| 2294 |
+
"learning_rate": 1.327041456712334e-05,
|
| 2295 |
+
"loss": 0.1281,
|
| 2296 |
+
"step": 7650
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 3.5474147924878277,
|
| 2300 |
+
"eval_loss": 0.15981091550942805,
|
| 2301 |
+
"eval_runtime": 60.5605,
|
| 2302 |
+
"eval_samples_per_second": 686.685,
|
| 2303 |
+
"eval_steps_per_second": 0.677,
|
| 2304 |
+
"step": 7650
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 3.570600510085787,
|
| 2308 |
+
"grad_norm": 0.04896661266684532,
|
| 2309 |
+
"learning_rate": 1.2736777516212266e-05,
|
| 2310 |
+
"loss": 0.1288,
|
| 2311 |
+
"step": 7700
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"epoch": 3.570600510085787,
|
| 2315 |
+
"eval_loss": 0.1599924400443614,
|
| 2316 |
+
"eval_runtime": 60.486,
|
| 2317 |
+
"eval_samples_per_second": 687.531,
|
| 2318 |
+
"eval_steps_per_second": 0.678,
|
| 2319 |
+
"step": 7700
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 3.5937862276837467,
|
| 2323 |
+
"grad_norm": 0.07458525151014328,
|
| 2324 |
+
"learning_rate": 1.2212521282287092e-05,
|
| 2325 |
+
"loss": 0.128,
|
| 2326 |
+
"step": 7750
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 3.5937862276837467,
|
| 2330 |
+
"eval_loss": 0.15936126278835275,
|
| 2331 |
+
"eval_runtime": 60.9341,
|
| 2332 |
+
"eval_samples_per_second": 682.475,
|
| 2333 |
+
"eval_steps_per_second": 0.673,
|
| 2334 |
+
"step": 7750
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 3.6169719452817066,
|
| 2338 |
+
"grad_norm": 0.04200127348303795,
|
| 2339 |
+
"learning_rate": 1.1697777844051105e-05,
|
| 2340 |
+
"loss": 0.1287,
|
| 2341 |
+
"step": 7800
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 3.6169719452817066,
|
| 2345 |
+
"eval_loss": 0.1603394617678833,
|
| 2346 |
+
"eval_runtime": 60.5155,
|
| 2347 |
+
"eval_samples_per_second": 687.195,
|
| 2348 |
+
"eval_steps_per_second": 0.678,
|
| 2349 |
+
"step": 7800
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 3.640157662879666,
|
| 2353 |
+
"grad_norm": 0.06712640821933746,
|
| 2354 |
+
"learning_rate": 1.1192676785412154e-05,
|
| 2355 |
+
"loss": 0.1291,
|
| 2356 |
+
"step": 7850
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 3.640157662879666,
|
| 2360 |
+
"eval_loss": 0.15920067938345067,
|
| 2361 |
+
"eval_runtime": 60.0225,
|
| 2362 |
+
"eval_samples_per_second": 692.84,
|
| 2363 |
+
"eval_steps_per_second": 0.683,
|
| 2364 |
+
"step": 7850
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"epoch": 3.6633433804776256,
|
| 2368 |
+
"grad_norm": 0.049462996423244476,
|
| 2369 |
+
"learning_rate": 1.0697345262860636e-05,
|
| 2370 |
+
"loss": 0.1287,
|
| 2371 |
+
"step": 7900
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 3.6633433804776256,
|
| 2375 |
+
"eval_loss": 0.15964593569874527,
|
| 2376 |
+
"eval_runtime": 60.1965,
|
| 2377 |
+
"eval_samples_per_second": 690.837,
|
| 2378 |
+
"eval_steps_per_second": 0.681,
|
| 2379 |
+
"step": 7900
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 3.6865290980755856,
|
| 2383 |
+
"grad_norm": 0.05148932337760925,
|
| 2384 |
+
"learning_rate": 1.021190797345839e-05,
|
| 2385 |
+
"loss": 0.1283,
|
| 2386 |
+
"step": 7950
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 3.6865290980755856,
|
| 2390 |
+
"eval_loss": 0.15903354419354673,
|
| 2391 |
+
"eval_runtime": 60.0507,
|
| 2392 |
+
"eval_samples_per_second": 692.515,
|
| 2393 |
+
"eval_steps_per_second": 0.683,
|
| 2394 |
+
"step": 7950
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 3.709714815673545,
|
| 2398 |
+
"grad_norm": 0.05164024233818054,
|
| 2399 |
+
"learning_rate": 9.73648712344707e-06,
|
| 2400 |
+
"loss": 0.128,
|
| 2401 |
+
"step": 8000
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 3.709714815673545,
|
| 2405 |
+
"eval_loss": 0.15835035051131605,
|
| 2406 |
+
"eval_runtime": 60.5739,
|
| 2407 |
+
"eval_samples_per_second": 686.533,
|
| 2408 |
+
"eval_steps_per_second": 0.677,
|
| 2409 |
+
"step": 8000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 3.7329005332715046,
|
| 2413 |
+
"grad_norm": 0.04926716163754463,
|
| 2414 |
+
"learning_rate": 9.271202397483215e-06,
|
| 2415 |
+
"loss": 0.1276,
|
| 2416 |
+
"step": 8050
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 3.7329005332715046,
|
| 2420 |
+
"eval_loss": 0.160225615529793,
|
| 2421 |
+
"eval_runtime": 60.4555,
|
| 2422 |
+
"eval_samples_per_second": 687.878,
|
| 2423 |
+
"eval_steps_per_second": 0.678,
|
| 2424 |
+
"step": 8050
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 3.7560862508694646,
|
| 2428 |
+
"grad_norm": 0.04355842247605324,
|
| 2429 |
+
"learning_rate": 8.816170928508365e-06,
|
| 2430 |
+
"loss": 0.1287,
|
| 2431 |
+
"step": 8100
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 3.7560862508694646,
|
| 2435 |
+
"eval_loss": 0.1601867779420742,
|
| 2436 |
+
"eval_runtime": 60.7386,
|
| 2437 |
+
"eval_samples_per_second": 684.672,
|
| 2438 |
+
"eval_steps_per_second": 0.675,
|
| 2439 |
+
"step": 8100
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 3.779271968467424,
|
| 2443 |
+
"grad_norm": 0.039105553179979324,
|
| 2444 |
+
"learning_rate": 8.371507268261437e-06,
|
| 2445 |
+
"loss": 0.1306,
|
| 2446 |
+
"step": 8150
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 3.779271968467424,
|
| 2450 |
+
"eval_loss": 0.15946348937187382,
|
| 2451 |
+
"eval_runtime": 60.9253,
|
| 2452 |
+
"eval_samples_per_second": 682.574,
|
| 2453 |
+
"eval_steps_per_second": 0.673,
|
| 2454 |
+
"step": 8150
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 3.8024576860653836,
|
| 2458 |
+
"grad_norm": 0.04452899843454361,
|
| 2459 |
+
"learning_rate": 7.937323358440935e-06,
|
| 2460 |
+
"loss": 0.1286,
|
| 2461 |
+
"step": 8200
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 3.8024576860653836,
|
| 2465 |
+
"eval_loss": 0.15871429728364056,
|
| 2466 |
+
"eval_runtime": 60.2776,
|
| 2467 |
+
"eval_samples_per_second": 689.908,
|
| 2468 |
+
"eval_steps_per_second": 0.68,
|
| 2469 |
+
"step": 8200
|
| 2470 |
+
},
|
| 2471 |
+
{
|
| 2472 |
+
"epoch": 3.8256434036633435,
|
| 2473 |
+
"grad_norm": 0.043075498193502426,
|
| 2474 |
+
"learning_rate": 7.513728502524286e-06,
|
| 2475 |
+
"loss": 0.1292,
|
| 2476 |
+
"step": 8250
|
| 2477 |
+
},
|
| 2478 |
+
{
|
| 2479 |
+
"epoch": 3.8256434036633435,
|
| 2480 |
+
"eval_loss": 0.1592580359542711,
|
| 2481 |
+
"eval_runtime": 60.7244,
|
| 2482 |
+
"eval_samples_per_second": 684.832,
|
| 2483 |
+
"eval_steps_per_second": 0.675,
|
| 2484 |
+
"step": 8250
|
| 2485 |
+
},
|
| 2486 |
+
{
|
| 2487 |
+
"epoch": 3.848829121261303,
|
| 2488 |
+
"grad_norm": 0.05848800390958786,
|
| 2489 |
+
"learning_rate": 7.100829338251147e-06,
|
| 2490 |
+
"loss": 0.1275,
|
| 2491 |
+
"step": 8300
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 3.848829121261303,
|
| 2495 |
+
"eval_loss": 0.15895083163665807,
|
| 2496 |
+
"eval_runtime": 60.3677,
|
| 2497 |
+
"eval_samples_per_second": 688.878,
|
| 2498 |
+
"eval_steps_per_second": 0.679,
|
| 2499 |
+
"step": 8300
|
| 2500 |
+
},
|
| 2501 |
+
{
|
| 2502 |
+
"epoch": 3.8720148388592626,
|
| 2503 |
+
"grad_norm": 0.04980336129665375,
|
| 2504 |
+
"learning_rate": 6.698729810778065e-06,
|
| 2505 |
+
"loss": 0.1277,
|
| 2506 |
+
"step": 8350
|
| 2507 |
+
},
|
| 2508 |
+
{
|
| 2509 |
+
"epoch": 3.8720148388592626,
|
| 2510 |
+
"eval_loss": 0.16002303550437,
|
| 2511 |
+
"eval_runtime": 60.2742,
|
| 2512 |
+
"eval_samples_per_second": 689.947,
|
| 2513 |
+
"eval_steps_per_second": 0.68,
|
| 2514 |
+
"step": 8350
|
| 2515 |
+
},
|
| 2516 |
+
{
|
| 2517 |
+
"epoch": 3.8952005564572225,
|
| 2518 |
+
"grad_norm": 0.057385146617889404,
|
| 2519 |
+
"learning_rate": 6.3075311465107535e-06,
|
| 2520 |
+
"loss": 0.129,
|
| 2521 |
+
"step": 8400
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 3.8952005564572225,
|
| 2525 |
+
"eval_loss": 0.1601535826416112,
|
| 2526 |
+
"eval_runtime": 60.4053,
|
| 2527 |
+
"eval_samples_per_second": 688.45,
|
| 2528 |
+
"eval_steps_per_second": 0.679,
|
| 2529 |
+
"step": 8400
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 3.918386274055182,
|
| 2533 |
+
"grad_norm": 0.045788682997226715,
|
| 2534 |
+
"learning_rate": 5.927331827620903e-06,
|
| 2535 |
+
"loss": 0.1286,
|
| 2536 |
+
"step": 8450
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 3.918386274055182,
|
| 2540 |
+
"eval_loss": 0.15926720973175468,
|
| 2541 |
+
"eval_runtime": 60.6783,
|
| 2542 |
+
"eval_samples_per_second": 685.352,
|
| 2543 |
+
"eval_steps_per_second": 0.676,
|
| 2544 |
+
"step": 8450
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 3.9415719916531415,
|
| 2548 |
+
"grad_norm": 0.045575451105833054,
|
| 2549 |
+
"learning_rate": 5.558227567253832e-06,
|
| 2550 |
+
"loss": 0.1281,
|
| 2551 |
+
"step": 8500
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 3.9415719916531415,
|
| 2555 |
+
"eval_loss": 0.16032033338606583,
|
| 2556 |
+
"eval_runtime": 60.4563,
|
| 2557 |
+
"eval_samples_per_second": 687.868,
|
| 2558 |
+
"eval_steps_per_second": 0.678,
|
| 2559 |
+
"step": 8500
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 3.964757709251101,
|
| 2563 |
+
"grad_norm": 0.034972067922353745,
|
| 2564 |
+
"learning_rate": 5.200311285433213e-06,
|
| 2565 |
+
"loss": 0.1285,
|
| 2566 |
+
"step": 8550
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 3.964757709251101,
|
| 2570 |
+
"eval_loss": 0.1590997571686103,
|
| 2571 |
+
"eval_runtime": 60.7642,
|
| 2572 |
+
"eval_samples_per_second": 684.384,
|
| 2573 |
+
"eval_steps_per_second": 0.675,
|
| 2574 |
+
"step": 8550
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 3.987943426849061,
|
| 2578 |
+
"grad_norm": 0.05060684680938721,
|
| 2579 |
+
"learning_rate": 4.853673085668947e-06,
|
| 2580 |
+
"loss": 0.1293,
|
| 2581 |
+
"step": 8600
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 3.987943426849061,
|
| 2585 |
+
"eval_loss": 0.15924322809570868,
|
| 2586 |
+
"eval_runtime": 60.0799,
|
| 2587 |
+
"eval_samples_per_second": 692.178,
|
| 2588 |
+
"eval_steps_per_second": 0.682,
|
| 2589 |
+
"step": 8600
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 4.011129144447021,
|
| 2593 |
+
"grad_norm": 0.04898017644882202,
|
| 2594 |
+
"learning_rate": 4.5184002322740785e-06,
|
| 2595 |
+
"loss": 0.1283,
|
| 2596 |
+
"step": 8650
|
| 2597 |
+
},
|
| 2598 |
+
{
|
| 2599 |
+
"epoch": 4.011129144447021,
|
| 2600 |
+
"eval_loss": 0.1587491140112498,
|
| 2601 |
+
"eval_runtime": 60.6393,
|
| 2602 |
+
"eval_samples_per_second": 685.793,
|
| 2603 |
+
"eval_steps_per_second": 0.676,
|
| 2604 |
+
"step": 8650
|
| 2605 |
+
},
|
| 2606 |
+
{
|
| 2607 |
+
"epoch": 4.0343148620449805,
|
| 2608 |
+
"grad_norm": 0.058361586183309555,
|
| 2609 |
+
"learning_rate": 4.19457712839652e-06,
|
| 2610 |
+
"loss": 0.1277,
|
| 2611 |
+
"step": 8700
|
| 2612 |
+
},
|
| 2613 |
+
{
|
| 2614 |
+
"epoch": 4.0343148620449805,
|
| 2615 |
+
"eval_loss": 0.1597737118597627,
|
| 2616 |
+
"eval_runtime": 61.5486,
|
| 2617 |
+
"eval_samples_per_second": 675.661,
|
| 2618 |
+
"eval_steps_per_second": 0.666,
|
| 2619 |
+
"step": 8700
|
| 2620 |
+
},
|
| 2621 |
+
{
|
| 2622 |
+
"epoch": 4.05750057964294,
|
| 2623 |
+
"grad_norm": 0.05138258635997772,
|
| 2624 |
+
"learning_rate": 3.8822852947709375e-06,
|
| 2625 |
+
"loss": 0.1283,
|
| 2626 |
+
"step": 8750
|
| 2627 |
+
},
|
| 2628 |
+
{
|
| 2629 |
+
"epoch": 4.05750057964294,
|
| 2630 |
+
"eval_loss": 0.15985116115580386,
|
| 2631 |
+
"eval_runtime": 60.5634,
|
| 2632 |
+
"eval_samples_per_second": 686.652,
|
| 2633 |
+
"eval_steps_per_second": 0.677,
|
| 2634 |
+
"step": 8750
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 4.0806862972408995,
|
| 2638 |
+
"grad_norm": 0.0461881086230278,
|
| 2639 |
+
"learning_rate": 3.581603349196372e-06,
|
| 2640 |
+
"loss": 0.1288,
|
| 2641 |
+
"step": 8800
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 4.0806862972408995,
|
| 2645 |
+
"eval_loss": 0.15788726458515429,
|
| 2646 |
+
"eval_runtime": 60.6057,
|
| 2647 |
+
"eval_samples_per_second": 686.173,
|
| 2648 |
+
"eval_steps_per_second": 0.677,
|
| 2649 |
+
"step": 8800
|
| 2650 |
+
}
|
| 2651 |
+
],
|
| 2652 |
+
"logging_steps": 50,
|
| 2653 |
+
"max_steps": 10000,
|
| 2654 |
+
"num_input_tokens_seen": 0,
|
| 2655 |
+
"num_train_epochs": 5,
|
| 2656 |
+
"save_steps": 50,
|
| 2657 |
+
"total_flos": 2.0443186999800627e+17,
|
| 2658 |
+
"train_batch_size": 1024,
|
| 2659 |
+
"trial_name": null,
|
| 2660 |
+
"trial_params": null
|
| 2661 |
+
}
|
checkpoint-8800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3df77597c69632ab7d5a5d7987fd817d663727a57d77ca069bc92a72238da772
|
| 3 |
+
size 5393
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"GPNRoFormerForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"aux_features_vocab_size": 5,
|
| 7 |
+
"embedding_size": 768,
|
| 8 |
+
"group_tokens": 1,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 1536,
|
| 16 |
+
"model_type": "GPNRoFormer",
|
| 17 |
+
"n_aux_features": 445,
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 2,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"rotary_value": false,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.40.2",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 6
|
| 27 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.637143519591931,
|
| 3 |
+
"eval_loss": 0.15925397718542236,
|
| 4 |
+
"eval_runtime": 61.5353,
|
| 5 |
+
"eval_samples_per_second": 675.807,
|
| 6 |
+
"eval_steps_per_second": 0.666,
|
| 7 |
+
"perplexity": 1.1726357315864648
|
| 8 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0e79347487c61d8217106a3f8f05bbf42d7c2038dab7a7a461077975e6acff9
|
| 3 |
+
size 59497084
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.637143519591931,
|
| 3 |
+
"total_flos": 2.3231400526217216e+17,
|
| 4 |
+
"train_loss": 0.13326009378433226,
|
| 5 |
+
"train_runtime": 41368.3669,
|
| 6 |
+
"train_samples_per_second": 495.064,
|
| 7 |
+
"train_steps_per_second": 0.242
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3030 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.15788726458515429,
|
| 3 |
+
"best_model_checkpoint": "checkpoints/checkpoint-8800",
|
| 4 |
+
"epoch": 4.637143519591931,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 10000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.023185717597959656,
|
| 13 |
+
"grad_norm": 0.16052097082138062,
|
| 14 |
+
"learning_rate": 5e-05,
|
| 15 |
+
"loss": 0.6225,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.023185717597959656,
|
| 20 |
+
"eval_loss": 0.1987911110084725,
|
| 21 |
+
"eval_runtime": 63.5433,
|
| 22 |
+
"eval_samples_per_second": 654.451,
|
| 23 |
+
"eval_steps_per_second": 0.645,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.04637143519591931,
|
| 28 |
+
"grad_norm": 0.09532159566879272,
|
| 29 |
+
"learning_rate": 0.0001,
|
| 30 |
+
"loss": 0.1508,
|
| 31 |
+
"step": 100
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.04637143519591931,
|
| 35 |
+
"eval_loss": 0.18357936446787168,
|
| 36 |
+
"eval_runtime": 60.9844,
|
| 37 |
+
"eval_samples_per_second": 681.912,
|
| 38 |
+
"eval_steps_per_second": 0.672,
|
| 39 |
+
"step": 100
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06955715279387897,
|
| 43 |
+
"grad_norm": 0.24056212604045868,
|
| 44 |
+
"learning_rate": 9.999370638369377e-05,
|
| 45 |
+
"loss": 0.1449,
|
| 46 |
+
"step": 150
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.06955715279387897,
|
| 50 |
+
"eval_loss": 0.17892658896642444,
|
| 51 |
+
"eval_runtime": 60.7834,
|
| 52 |
+
"eval_samples_per_second": 684.167,
|
| 53 |
+
"eval_steps_per_second": 0.675,
|
| 54 |
+
"step": 150
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09274287039183862,
|
| 58 |
+
"grad_norm": 0.09350813180208206,
|
| 59 |
+
"learning_rate": 9.997482711915927e-05,
|
| 60 |
+
"loss": 0.1421,
|
| 61 |
+
"step": 200
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.09274287039183862,
|
| 65 |
+
"eval_loss": 0.17624869869175752,
|
| 66 |
+
"eval_runtime": 60.3826,
|
| 67 |
+
"eval_samples_per_second": 688.708,
|
| 68 |
+
"eval_steps_per_second": 0.679,
|
| 69 |
+
"step": 200
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.11592858798979828,
|
| 73 |
+
"grad_norm": 0.12230529636144638,
|
| 74 |
+
"learning_rate": 9.99433669591504e-05,
|
| 75 |
+
"loss": 0.141,
|
| 76 |
+
"step": 250
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.11592858798979828,
|
| 80 |
+
"eval_loss": 0.17641382363047173,
|
| 81 |
+
"eval_runtime": 60.4169,
|
| 82 |
+
"eval_samples_per_second": 688.317,
|
| 83 |
+
"eval_steps_per_second": 0.679,
|
| 84 |
+
"step": 250
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.13911430558775795,
|
| 88 |
+
"grad_norm": 0.14592748880386353,
|
| 89 |
+
"learning_rate": 9.989933382359422e-05,
|
| 90 |
+
"loss": 0.1397,
|
| 91 |
+
"step": 300
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.13911430558775795,
|
| 95 |
+
"eval_loss": 0.17552215792639078,
|
| 96 |
+
"eval_runtime": 61.6101,
|
| 97 |
+
"eval_samples_per_second": 674.987,
|
| 98 |
+
"eval_steps_per_second": 0.665,
|
| 99 |
+
"step": 300
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.1623000231857176,
|
| 103 |
+
"grad_norm": 0.10219988226890564,
|
| 104 |
+
"learning_rate": 9.984273879759713e-05,
|
| 105 |
+
"loss": 0.1393,
|
| 106 |
+
"step": 350
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.1623000231857176,
|
| 110 |
+
"eval_loss": 0.17414749172793012,
|
| 111 |
+
"eval_runtime": 61.4962,
|
| 112 |
+
"eval_samples_per_second": 676.237,
|
| 113 |
+
"eval_steps_per_second": 0.667,
|
| 114 |
+
"step": 350
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.18548574078367724,
|
| 118 |
+
"grad_norm": 0.12338168174028397,
|
| 119 |
+
"learning_rate": 9.977359612865423e-05,
|
| 120 |
+
"loss": 0.1388,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.18548574078367724,
|
| 125 |
+
"eval_loss": 0.17378012638412807,
|
| 126 |
+
"eval_runtime": 61.0462,
|
| 127 |
+
"eval_samples_per_second": 681.221,
|
| 128 |
+
"eval_steps_per_second": 0.672,
|
| 129 |
+
"step": 400
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.20867145838163692,
|
| 133 |
+
"grad_norm": 0.09479879587888718,
|
| 134 |
+
"learning_rate": 9.969192322306271e-05,
|
| 135 |
+
"loss": 0.1394,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.20867145838163692,
|
| 140 |
+
"eval_loss": 0.17252362204688398,
|
| 141 |
+
"eval_runtime": 60.963,
|
| 142 |
+
"eval_samples_per_second": 682.151,
|
| 143 |
+
"eval_steps_per_second": 0.673,
|
| 144 |
+
"step": 450
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.23185717597959657,
|
| 148 |
+
"grad_norm": 0.1108623668551445,
|
| 149 |
+
"learning_rate": 9.959774064153977e-05,
|
| 150 |
+
"loss": 0.1383,
|
| 151 |
+
"step": 500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.23185717597959657,
|
| 155 |
+
"eval_loss": 0.17298176916843877,
|
| 156 |
+
"eval_runtime": 60.6546,
|
| 157 |
+
"eval_samples_per_second": 685.62,
|
| 158 |
+
"eval_steps_per_second": 0.676,
|
| 159 |
+
"step": 500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.2550428935775562,
|
| 163 |
+
"grad_norm": 0.0725204199552536,
|
| 164 |
+
"learning_rate": 9.949107209404665e-05,
|
| 165 |
+
"loss": 0.1376,
|
| 166 |
+
"step": 550
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.2550428935775562,
|
| 170 |
+
"eval_loss": 0.17165218165539878,
|
| 171 |
+
"eval_runtime": 59.4888,
|
| 172 |
+
"eval_samples_per_second": 699.056,
|
| 173 |
+
"eval_steps_per_second": 0.689,
|
| 174 |
+
"step": 550
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.2782286111755159,
|
| 178 |
+
"grad_norm": 0.0955963134765625,
|
| 179 |
+
"learning_rate": 9.937194443381972e-05,
|
| 180 |
+
"loss": 0.1372,
|
| 181 |
+
"step": 600
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.2782286111755159,
|
| 185 |
+
"eval_loss": 0.17077083113718278,
|
| 186 |
+
"eval_runtime": 60.6021,
|
| 187 |
+
"eval_samples_per_second": 686.214,
|
| 188 |
+
"eval_steps_per_second": 0.677,
|
| 189 |
+
"step": 600
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.3014143287734755,
|
| 193 |
+
"grad_norm": 0.18736732006072998,
|
| 194 |
+
"learning_rate": 9.924038765061042e-05,
|
| 195 |
+
"loss": 0.1361,
|
| 196 |
+
"step": 650
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.3014143287734755,
|
| 200 |
+
"eval_loss": 0.1727738813183492,
|
| 201 |
+
"eval_runtime": 60.3343,
|
| 202 |
+
"eval_samples_per_second": 689.259,
|
| 203 |
+
"eval_steps_per_second": 0.68,
|
| 204 |
+
"step": 650
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.3246000463714352,
|
| 208 |
+
"grad_norm": 0.09572151303291321,
|
| 209 |
+
"learning_rate": 9.909643486313533e-05,
|
| 210 |
+
"loss": 0.1362,
|
| 211 |
+
"step": 700
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.3246000463714352,
|
| 215 |
+
"eval_loss": 0.17145407115151273,
|
| 216 |
+
"eval_runtime": 60.2732,
|
| 217 |
+
"eval_samples_per_second": 689.959,
|
| 218 |
+
"eval_steps_per_second": 0.68,
|
| 219 |
+
"step": 700
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.34778576396939487,
|
| 223 |
+
"grad_norm": 0.07214252650737762,
|
| 224 |
+
"learning_rate": 9.894012231073894e-05,
|
| 225 |
+
"loss": 0.1364,
|
| 226 |
+
"step": 750
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.34778576396939487,
|
| 230 |
+
"eval_loss": 0.17133199408489355,
|
| 231 |
+
"eval_runtime": 60.0148,
|
| 232 |
+
"eval_samples_per_second": 692.929,
|
| 233 |
+
"eval_steps_per_second": 0.683,
|
| 234 |
+
"step": 750
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.3709714815673545,
|
| 238 |
+
"grad_norm": 0.18224318325519562,
|
| 239 |
+
"learning_rate": 9.877148934427037e-05,
|
| 240 |
+
"loss": 0.1356,
|
| 241 |
+
"step": 800
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3709714815673545,
|
| 245 |
+
"eval_loss": 0.16949569222888886,
|
| 246 |
+
"eval_runtime": 60.1491,
|
| 247 |
+
"eval_samples_per_second": 691.382,
|
| 248 |
+
"eval_steps_per_second": 0.682,
|
| 249 |
+
"step": 800
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.39415719916531416,
|
| 253 |
+
"grad_norm": 0.06306415796279907,
|
| 254 |
+
"learning_rate": 9.859057841617709e-05,
|
| 255 |
+
"loss": 0.1353,
|
| 256 |
+
"step": 850
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.39415719916531416,
|
| 260 |
+
"eval_loss": 0.1686952690798172,
|
| 261 |
+
"eval_runtime": 60.6447,
|
| 262 |
+
"eval_samples_per_second": 685.731,
|
| 263 |
+
"eval_steps_per_second": 0.676,
|
| 264 |
+
"step": 850
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.41734291676327384,
|
| 268 |
+
"grad_norm": 0.10090287029743195,
|
| 269 |
+
"learning_rate": 9.839743506981782e-05,
|
| 270 |
+
"loss": 0.1361,
|
| 271 |
+
"step": 900
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.41734291676327384,
|
| 275 |
+
"eval_loss": 0.17026100034088926,
|
| 276 |
+
"eval_runtime": 61.5224,
|
| 277 |
+
"eval_samples_per_second": 675.949,
|
| 278 |
+
"eval_steps_per_second": 0.666,
|
| 279 |
+
"step": 900
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.44052863436123346,
|
| 283 |
+
"grad_norm": 0.10061236470937729,
|
| 284 |
+
"learning_rate": 9.819210792799712e-05,
|
| 285 |
+
"loss": 0.1354,
|
| 286 |
+
"step": 950
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.44052863436123346,
|
| 290 |
+
"eval_loss": 0.16971544565694113,
|
| 291 |
+
"eval_runtime": 60.488,
|
| 292 |
+
"eval_samples_per_second": 687.508,
|
| 293 |
+
"eval_steps_per_second": 0.678,
|
| 294 |
+
"step": 950
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.46371435195919314,
|
| 298 |
+
"grad_norm": 0.06525534391403198,
|
| 299 |
+
"learning_rate": 9.797464868072488e-05,
|
| 300 |
+
"loss": 0.1352,
|
| 301 |
+
"step": 1000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.46371435195919314,
|
| 305 |
+
"eval_loss": 0.16946903195393553,
|
| 306 |
+
"eval_runtime": 61.3558,
|
| 307 |
+
"eval_samples_per_second": 677.784,
|
| 308 |
+
"eval_steps_per_second": 0.668,
|
| 309 |
+
"step": 1000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.4869000695571528,
|
| 313 |
+
"grad_norm": 0.06269507855176926,
|
| 314 |
+
"learning_rate": 9.77451120722037e-05,
|
| 315 |
+
"loss": 0.1335,
|
| 316 |
+
"step": 1050
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.4869000695571528,
|
| 320 |
+
"eval_loss": 0.16825352947444114,
|
| 321 |
+
"eval_runtime": 60.1971,
|
| 322 |
+
"eval_samples_per_second": 690.831,
|
| 323 |
+
"eval_steps_per_second": 0.681,
|
| 324 |
+
"step": 1050
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5100857871551124,
|
| 328 |
+
"grad_norm": 0.08187470585107803,
|
| 329 |
+
"learning_rate": 9.750355588704727e-05,
|
| 330 |
+
"loss": 0.1327,
|
| 331 |
+
"step": 1100
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5100857871551124,
|
| 335 |
+
"eval_loss": 0.16861737282523587,
|
| 336 |
+
"eval_runtime": 59.5715,
|
| 337 |
+
"eval_samples_per_second": 698.085,
|
| 338 |
+
"eval_steps_per_second": 0.688,
|
| 339 |
+
"step": 1100
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5332715047530721,
|
| 343 |
+
"grad_norm": 0.06607680767774582,
|
| 344 |
+
"learning_rate": 9.725004093573342e-05,
|
| 345 |
+
"loss": 0.1337,
|
| 346 |
+
"step": 1150
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5332715047530721,
|
| 350 |
+
"eval_loss": 0.1692070748762034,
|
| 351 |
+
"eval_runtime": 60.0637,
|
| 352 |
+
"eval_samples_per_second": 692.364,
|
| 353 |
+
"eval_steps_per_second": 0.683,
|
| 354 |
+
"step": 1150
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.5564572223510318,
|
| 358 |
+
"grad_norm": 0.09759815782308578,
|
| 359 |
+
"learning_rate": 9.698463103929542e-05,
|
| 360 |
+
"loss": 0.134,
|
| 361 |
+
"step": 1200
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5564572223510318,
|
| 365 |
+
"eval_loss": 0.16649228385381692,
|
| 366 |
+
"eval_runtime": 60.4305,
|
| 367 |
+
"eval_samples_per_second": 688.162,
|
| 368 |
+
"eval_steps_per_second": 0.678,
|
| 369 |
+
"step": 1200
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.5796429399489914,
|
| 373 |
+
"grad_norm": 0.10353852063417435,
|
| 374 |
+
"learning_rate": 9.670739301325534e-05,
|
| 375 |
+
"loss": 0.1341,
|
| 376 |
+
"step": 1250
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.5796429399489914,
|
| 380 |
+
"eval_loss": 0.16802514322459206,
|
| 381 |
+
"eval_runtime": 60.0955,
|
| 382 |
+
"eval_samples_per_second": 691.999,
|
| 383 |
+
"eval_steps_per_second": 0.682,
|
| 384 |
+
"step": 1250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.602828657546951,
|
| 388 |
+
"grad_norm": 0.11834366619586945,
|
| 389 |
+
"learning_rate": 9.641839665080363e-05,
|
| 390 |
+
"loss": 0.1347,
|
| 391 |
+
"step": 1300
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.602828657546951,
|
| 395 |
+
"eval_loss": 0.1672302417427292,
|
| 396 |
+
"eval_runtime": 60.3484,
|
| 397 |
+
"eval_samples_per_second": 689.098,
|
| 398 |
+
"eval_steps_per_second": 0.679,
|
| 399 |
+
"step": 1300
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.6260143751449108,
|
| 403 |
+
"grad_norm": 0.06963012367486954,
|
| 404 |
+
"learning_rate": 9.611771470522908e-05,
|
| 405 |
+
"loss": 0.1335,
|
| 406 |
+
"step": 1350
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.6260143751449108,
|
| 410 |
+
"eval_loss": 0.16607839684977216,
|
| 411 |
+
"eval_runtime": 60.3308,
|
| 412 |
+
"eval_samples_per_second": 689.3,
|
| 413 |
+
"eval_steps_per_second": 0.68,
|
| 414 |
+
"step": 1350
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.6492000927428704,
|
| 418 |
+
"grad_norm": 0.06842990219593048,
|
| 419 |
+
"learning_rate": 9.580542287160348e-05,
|
| 420 |
+
"loss": 0.1338,
|
| 421 |
+
"step": 1400
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.6492000927428704,
|
| 425 |
+
"eval_loss": 0.16628812684035693,
|
| 426 |
+
"eval_runtime": 59.9335,
|
| 427 |
+
"eval_samples_per_second": 693.87,
|
| 428 |
+
"eval_steps_per_second": 0.684,
|
| 429 |
+
"step": 1400
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.67238581034083,
|
| 433 |
+
"grad_norm": 0.07053674757480621,
|
| 434 |
+
"learning_rate": 9.548159976772592e-05,
|
| 435 |
+
"loss": 0.1335,
|
| 436 |
+
"step": 1450
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.67238581034083,
|
| 440 |
+
"eval_loss": 0.16696060882262428,
|
| 441 |
+
"eval_runtime": 59.8079,
|
| 442 |
+
"eval_samples_per_second": 695.326,
|
| 443 |
+
"eval_steps_per_second": 0.686,
|
| 444 |
+
"step": 1450
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6955715279387897,
|
| 448 |
+
"grad_norm": 0.09175281971693039,
|
| 449 |
+
"learning_rate": 9.514632691433107e-05,
|
| 450 |
+
"loss": 0.1332,
|
| 451 |
+
"step": 1500
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6955715279387897,
|
| 455 |
+
"eval_loss": 0.16521949465081834,
|
| 456 |
+
"eval_runtime": 60.1856,
|
| 457 |
+
"eval_samples_per_second": 690.963,
|
| 458 |
+
"eval_steps_per_second": 0.681,
|
| 459 |
+
"step": 1500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.7187572455367494,
|
| 463 |
+
"grad_norm": 0.05836635082960129,
|
| 464 |
+
"learning_rate": 9.479968871456679e-05,
|
| 465 |
+
"loss": 0.1336,
|
| 466 |
+
"step": 1550
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.7187572455367494,
|
| 470 |
+
"eval_loss": 0.16626366255041727,
|
| 471 |
+
"eval_runtime": 60.6256,
|
| 472 |
+
"eval_samples_per_second": 685.948,
|
| 473 |
+
"eval_steps_per_second": 0.676,
|
| 474 |
+
"step": 1550
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.741942963134709,
|
| 478 |
+
"grad_norm": 0.07249301671981812,
|
| 479 |
+
"learning_rate": 9.444177243274618e-05,
|
| 480 |
+
"loss": 0.133,
|
| 481 |
+
"step": 1600
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.741942963134709,
|
| 485 |
+
"eval_loss": 0.1655649439629329,
|
| 486 |
+
"eval_runtime": 60.2447,
|
| 487 |
+
"eval_samples_per_second": 690.285,
|
| 488 |
+
"eval_steps_per_second": 0.681,
|
| 489 |
+
"step": 1600
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.7651286807326687,
|
| 493 |
+
"grad_norm": 0.07509302347898483,
|
| 494 |
+
"learning_rate": 9.407266817237911e-05,
|
| 495 |
+
"loss": 0.1332,
|
| 496 |
+
"step": 1650
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.7651286807326687,
|
| 500 |
+
"eval_loss": 0.16605371203296967,
|
| 501 |
+
"eval_runtime": 59.8196,
|
| 502 |
+
"eval_samples_per_second": 695.191,
|
| 503 |
+
"eval_steps_per_second": 0.685,
|
| 504 |
+
"step": 1650
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.7883143983306283,
|
| 508 |
+
"grad_norm": 0.07540406286716461,
|
| 509 |
+
"learning_rate": 9.369246885348926e-05,
|
| 510 |
+
"loss": 0.1327,
|
| 511 |
+
"step": 1700
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.7883143983306283,
|
| 515 |
+
"eval_loss": 0.16555590021301406,
|
| 516 |
+
"eval_runtime": 60.4119,
|
| 517 |
+
"eval_samples_per_second": 688.374,
|
| 518 |
+
"eval_steps_per_second": 0.679,
|
| 519 |
+
"step": 1700
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.811500115928588,
|
| 523 |
+
"grad_norm": 0.06061087176203728,
|
| 524 |
+
"learning_rate": 9.330127018922194e-05,
|
| 525 |
+
"loss": 0.1318,
|
| 526 |
+
"step": 1750
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.811500115928588,
|
| 530 |
+
"eval_loss": 0.16623179673527624,
|
| 531 |
+
"eval_runtime": 59.7807,
|
| 532 |
+
"eval_samples_per_second": 695.643,
|
| 533 |
+
"eval_steps_per_second": 0.686,
|
| 534 |
+
"step": 1750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.8346858335265477,
|
| 538 |
+
"grad_norm": 0.05577518790960312,
|
| 539 |
+
"learning_rate": 9.289917066174886e-05,
|
| 540 |
+
"loss": 0.1319,
|
| 541 |
+
"step": 1800
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.8346858335265477,
|
| 545 |
+
"eval_loss": 0.16519989030959317,
|
| 546 |
+
"eval_runtime": 60.1508,
|
| 547 |
+
"eval_samples_per_second": 691.363,
|
| 548 |
+
"eval_steps_per_second": 0.682,
|
| 549 |
+
"step": 1800
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8578715511245073,
|
| 553 |
+
"grad_norm": 0.06929640471935272,
|
| 554 |
+
"learning_rate": 9.248627149747573e-05,
|
| 555 |
+
"loss": 0.1337,
|
| 556 |
+
"step": 1850
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.8578715511245073,
|
| 560 |
+
"eval_loss": 0.16394849849125304,
|
| 561 |
+
"eval_runtime": 60.1044,
|
| 562 |
+
"eval_samples_per_second": 691.896,
|
| 563 |
+
"eval_steps_per_second": 0.682,
|
| 564 |
+
"step": 1850
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.8810572687224669,
|
| 568 |
+
"grad_norm": 0.07941466569900513,
|
| 569 |
+
"learning_rate": 9.206267664155907e-05,
|
| 570 |
+
"loss": 0.1324,
|
| 571 |
+
"step": 1900
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.8810572687224669,
|
| 575 |
+
"eval_loss": 0.1648257591054525,
|
| 576 |
+
"eval_runtime": 59.9818,
|
| 577 |
+
"eval_samples_per_second": 693.31,
|
| 578 |
+
"eval_steps_per_second": 0.684,
|
| 579 |
+
"step": 1900
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.9042429863204267,
|
| 583 |
+
"grad_norm": 0.09700328856706619,
|
| 584 |
+
"learning_rate": 9.162849273173857e-05,
|
| 585 |
+
"loss": 0.1334,
|
| 586 |
+
"step": 1950
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.9042429863204267,
|
| 590 |
+
"eval_loss": 0.16508159820082235,
|
| 591 |
+
"eval_runtime": 60.0956,
|
| 592 |
+
"eval_samples_per_second": 691.997,
|
| 593 |
+
"eval_steps_per_second": 0.682,
|
| 594 |
+
"step": 1950
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.9274287039183863,
|
| 598 |
+
"grad_norm": 0.09397923946380615,
|
| 599 |
+
"learning_rate": 9.118382907149165e-05,
|
| 600 |
+
"loss": 0.1317,
|
| 601 |
+
"step": 2000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.9274287039183863,
|
| 605 |
+
"eval_loss": 0.16377645660046958,
|
| 606 |
+
"eval_runtime": 60.471,
|
| 607 |
+
"eval_samples_per_second": 687.701,
|
| 608 |
+
"eval_steps_per_second": 0.678,
|
| 609 |
+
"step": 2000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.9506144215163459,
|
| 613 |
+
"grad_norm": 0.08097202330827713,
|
| 614 |
+
"learning_rate": 9.072879760251679e-05,
|
| 615 |
+
"loss": 0.1324,
|
| 616 |
+
"step": 2050
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.9506144215163459,
|
| 620 |
+
"eval_loss": 0.16491611914973717,
|
| 621 |
+
"eval_runtime": 60.6678,
|
| 622 |
+
"eval_samples_per_second": 685.471,
|
| 623 |
+
"eval_steps_per_second": 0.676,
|
| 624 |
+
"step": 2050
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.9738001391143056,
|
| 628 |
+
"grad_norm": 0.08455361425876617,
|
| 629 |
+
"learning_rate": 9.026351287655294e-05,
|
| 630 |
+
"loss": 0.1326,
|
| 631 |
+
"step": 2100
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.9738001391143056,
|
| 635 |
+
"eval_loss": 0.16602741997032858,
|
| 636 |
+
"eval_runtime": 60.5593,
|
| 637 |
+
"eval_samples_per_second": 686.698,
|
| 638 |
+
"eval_steps_per_second": 0.677,
|
| 639 |
+
"step": 2100
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9969858567122653,
|
| 643 |
+
"grad_norm": 0.056316621601581573,
|
| 644 |
+
"learning_rate": 8.978809202654162e-05,
|
| 645 |
+
"loss": 0.1326,
|
| 646 |
+
"step": 2150
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.9969858567122653,
|
| 650 |
+
"eval_loss": 0.1640188218462461,
|
| 651 |
+
"eval_runtime": 60.9228,
|
| 652 |
+
"eval_samples_per_second": 682.602,
|
| 653 |
+
"eval_steps_per_second": 0.673,
|
| 654 |
+
"step": 2150
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.0201715743102249,
|
| 658 |
+
"grad_norm": 0.06686601787805557,
|
| 659 |
+
"learning_rate": 8.930265473713938e-05,
|
| 660 |
+
"loss": 0.132,
|
| 661 |
+
"step": 2200
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.0201715743102249,
|
| 665 |
+
"eval_loss": 0.1652621257294944,
|
| 666 |
+
"eval_runtime": 60.883,
|
| 667 |
+
"eval_samples_per_second": 683.048,
|
| 668 |
+
"eval_steps_per_second": 0.673,
|
| 669 |
+
"step": 2200
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.0433572919081846,
|
| 673 |
+
"grad_norm": 0.040202509611845016,
|
| 674 |
+
"learning_rate": 8.880732321458784e-05,
|
| 675 |
+
"loss": 0.1319,
|
| 676 |
+
"step": 2250
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.0433572919081846,
|
| 680 |
+
"eval_loss": 0.1655291575008717,
|
| 681 |
+
"eval_runtime": 60.3109,
|
| 682 |
+
"eval_samples_per_second": 689.527,
|
| 683 |
+
"eval_steps_per_second": 0.68,
|
| 684 |
+
"step": 2250
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.0665430095061441,
|
| 688 |
+
"grad_norm": 0.0656428411602974,
|
| 689 |
+
"learning_rate": 8.83022221559489e-05,
|
| 690 |
+
"loss": 0.1326,
|
| 691 |
+
"step": 2300
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 1.0665430095061441,
|
| 695 |
+
"eval_loss": 0.16431572407087935,
|
| 696 |
+
"eval_runtime": 60.1036,
|
| 697 |
+
"eval_samples_per_second": 691.906,
|
| 698 |
+
"eval_steps_per_second": 0.682,
|
| 699 |
+
"step": 2300
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.0897287271041038,
|
| 703 |
+
"grad_norm": 0.06945247948169708,
|
| 704 |
+
"learning_rate": 8.778747871771292e-05,
|
| 705 |
+
"loss": 0.1321,
|
| 706 |
+
"step": 2350
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 1.0897287271041038,
|
| 710 |
+
"eval_loss": 0.16585482329987242,
|
| 711 |
+
"eval_runtime": 60.6263,
|
| 712 |
+
"eval_samples_per_second": 685.94,
|
| 713 |
+
"eval_steps_per_second": 0.676,
|
| 714 |
+
"step": 2350
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.1129144447020636,
|
| 718 |
+
"grad_norm": 0.0523492731153965,
|
| 719 |
+
"learning_rate": 8.726322248378775e-05,
|
| 720 |
+
"loss": 0.1317,
|
| 721 |
+
"step": 2400
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 1.1129144447020636,
|
| 725 |
+
"eval_loss": 0.16438524923736036,
|
| 726 |
+
"eval_runtime": 60.317,
|
| 727 |
+
"eval_samples_per_second": 689.457,
|
| 728 |
+
"eval_steps_per_second": 0.68,
|
| 729 |
+
"step": 2400
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.136100162300023,
|
| 733 |
+
"grad_norm": 0.07777334004640579,
|
| 734 |
+
"learning_rate": 8.672958543287666e-05,
|
| 735 |
+
"loss": 0.1322,
|
| 736 |
+
"step": 2450
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.136100162300023,
|
| 740 |
+
"eval_loss": 0.16509696053644565,
|
| 741 |
+
"eval_runtime": 60.26,
|
| 742 |
+
"eval_samples_per_second": 690.109,
|
| 743 |
+
"eval_steps_per_second": 0.68,
|
| 744 |
+
"step": 2450
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.1592858798979828,
|
| 748 |
+
"grad_norm": 0.06430637836456299,
|
| 749 |
+
"learning_rate": 8.618670190525352e-05,
|
| 750 |
+
"loss": 0.1325,
|
| 751 |
+
"step": 2500
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.1592858798979828,
|
| 755 |
+
"eval_loss": 0.1639541608445008,
|
| 756 |
+
"eval_runtime": 60.5314,
|
| 757 |
+
"eval_samples_per_second": 687.015,
|
| 758 |
+
"eval_steps_per_second": 0.677,
|
| 759 |
+
"step": 2500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.1824715974959426,
|
| 763 |
+
"grad_norm": 0.11194106936454773,
|
| 764 |
+
"learning_rate": 8.563470856894316e-05,
|
| 765 |
+
"loss": 0.1311,
|
| 766 |
+
"step": 2550
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.1824715974959426,
|
| 770 |
+
"eval_loss": 0.16260699934317355,
|
| 771 |
+
"eval_runtime": 60.3659,
|
| 772 |
+
"eval_samples_per_second": 688.899,
|
| 773 |
+
"eval_steps_per_second": 0.679,
|
| 774 |
+
"step": 2550
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.205657315093902,
|
| 778 |
+
"grad_norm": 0.06165901944041252,
|
| 779 |
+
"learning_rate": 8.507374438531607e-05,
|
| 780 |
+
"loss": 0.1323,
|
| 781 |
+
"step": 2600
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.205657315093902,
|
| 785 |
+
"eval_loss": 0.1626319663130242,
|
| 786 |
+
"eval_runtime": 59.9516,
|
| 787 |
+
"eval_samples_per_second": 693.66,
|
| 788 |
+
"eval_steps_per_second": 0.684,
|
| 789 |
+
"step": 2600
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.2288430326918618,
|
| 793 |
+
"grad_norm": 0.10654885321855545,
|
| 794 |
+
"learning_rate": 8.450395057410561e-05,
|
| 795 |
+
"loss": 0.1316,
|
| 796 |
+
"step": 2650
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.2288430326918618,
|
| 800 |
+
"eval_loss": 0.16393000041041636,
|
| 801 |
+
"eval_runtime": 59.576,
|
| 802 |
+
"eval_samples_per_second": 698.032,
|
| 803 |
+
"eval_steps_per_second": 0.688,
|
| 804 |
+
"step": 2650
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.2520287502898215,
|
| 808 |
+
"grad_norm": 0.04848140478134155,
|
| 809 |
+
"learning_rate": 8.392547057785661e-05,
|
| 810 |
+
"loss": 0.1314,
|
| 811 |
+
"step": 2700
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.2520287502898215,
|
| 815 |
+
"eval_loss": 0.16348152455768114,
|
| 816 |
+
"eval_runtime": 60.098,
|
| 817 |
+
"eval_samples_per_second": 691.97,
|
| 818 |
+
"eval_steps_per_second": 0.682,
|
| 819 |
+
"step": 2700
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.275214467887781,
|
| 823 |
+
"grad_norm": 0.0573604516685009,
|
| 824 |
+
"learning_rate": 8.333845002581458e-05,
|
| 825 |
+
"loss": 0.1314,
|
| 826 |
+
"step": 2750
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.275214467887781,
|
| 830 |
+
"eval_loss": 0.16364089140116167,
|
| 831 |
+
"eval_runtime": 60.1364,
|
| 832 |
+
"eval_samples_per_second": 691.528,
|
| 833 |
+
"eval_steps_per_second": 0.682,
|
| 834 |
+
"step": 2750
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.2984001854857408,
|
| 838 |
+
"grad_norm": 0.053159259259700775,
|
| 839 |
+
"learning_rate": 8.274303669726426e-05,
|
| 840 |
+
"loss": 0.131,
|
| 841 |
+
"step": 2800
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.2984001854857408,
|
| 845 |
+
"eval_loss": 0.16257415365129801,
|
| 846 |
+
"eval_runtime": 60.0025,
|
| 847 |
+
"eval_samples_per_second": 693.071,
|
| 848 |
+
"eval_steps_per_second": 0.683,
|
| 849 |
+
"step": 2800
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.3215859030837005,
|
| 853 |
+
"grad_norm": 0.09136148542165756,
|
| 854 |
+
"learning_rate": 8.213938048432697e-05,
|
| 855 |
+
"loss": 0.1313,
|
| 856 |
+
"step": 2850
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.3215859030837005,
|
| 860 |
+
"eval_loss": 0.16324665471619784,
|
| 861 |
+
"eval_runtime": 59.8429,
|
| 862 |
+
"eval_samples_per_second": 694.92,
|
| 863 |
+
"eval_steps_per_second": 0.685,
|
| 864 |
+
"step": 2850
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.34477162068166,
|
| 868 |
+
"grad_norm": 0.05825324356555939,
|
| 869 |
+
"learning_rate": 8.152763335422613e-05,
|
| 870 |
+
"loss": 0.1312,
|
| 871 |
+
"step": 2900
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.34477162068166,
|
| 875 |
+
"eval_loss": 0.16367374608121235,
|
| 876 |
+
"eval_runtime": 60.219,
|
| 877 |
+
"eval_samples_per_second": 690.579,
|
| 878 |
+
"eval_steps_per_second": 0.681,
|
| 879 |
+
"step": 2900
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.3679573382796197,
|
| 883 |
+
"grad_norm": 0.06379790604114532,
|
| 884 |
+
"learning_rate": 8.090794931103026e-05,
|
| 885 |
+
"loss": 0.1317,
|
| 886 |
+
"step": 2950
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.3679573382796197,
|
| 890 |
+
"eval_loss": 0.16400733758786312,
|
| 891 |
+
"eval_runtime": 59.9641,
|
| 892 |
+
"eval_samples_per_second": 693.515,
|
| 893 |
+
"eval_steps_per_second": 0.684,
|
| 894 |
+
"step": 2950
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.3911430558775795,
|
| 898 |
+
"grad_norm": 0.05361103266477585,
|
| 899 |
+
"learning_rate": 8.028048435688333e-05,
|
| 900 |
+
"loss": 0.1311,
|
| 901 |
+
"step": 3000
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 1.3911430558775795,
|
| 905 |
+
"eval_loss": 0.16210626991928834,
|
| 906 |
+
"eval_runtime": 59.5858,
|
| 907 |
+
"eval_samples_per_second": 697.919,
|
| 908 |
+
"eval_steps_per_second": 0.688,
|
| 909 |
+
"step": 3000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.414328773475539,
|
| 913 |
+
"grad_norm": 0.04593402519822121,
|
| 914 |
+
"learning_rate": 7.964539645273204e-05,
|
| 915 |
+
"loss": 0.1304,
|
| 916 |
+
"step": 3050
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 1.414328773475539,
|
| 920 |
+
"eval_loss": 0.163067463275087,
|
| 921 |
+
"eval_runtime": 60.098,
|
| 922 |
+
"eval_samples_per_second": 691.97,
|
| 923 |
+
"eval_steps_per_second": 0.682,
|
| 924 |
+
"step": 3050
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.4375144910734987,
|
| 928 |
+
"grad_norm": 0.057480327785015106,
|
| 929 |
+
"learning_rate": 7.900284547855991e-05,
|
| 930 |
+
"loss": 0.1307,
|
| 931 |
+
"step": 3100
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 1.4375144910734987,
|
| 935 |
+
"eval_loss": 0.16243572043734797,
|
| 936 |
+
"eval_runtime": 59.5674,
|
| 937 |
+
"eval_samples_per_second": 698.133,
|
| 938 |
+
"eval_steps_per_second": 0.688,
|
| 939 |
+
"step": 3100
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.4607002086714584,
|
| 943 |
+
"grad_norm": 0.08223798871040344,
|
| 944 |
+
"learning_rate": 7.835299319313853e-05,
|
| 945 |
+
"loss": 0.1315,
|
| 946 |
+
"step": 3150
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.4607002086714584,
|
| 950 |
+
"eval_loss": 0.1641944734489707,
|
| 951 |
+
"eval_runtime": 59.5423,
|
| 952 |
+
"eval_samples_per_second": 698.428,
|
| 953 |
+
"eval_steps_per_second": 0.689,
|
| 954 |
+
"step": 3150
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.483885926269418,
|
| 958 |
+
"grad_norm": 0.09742949903011322,
|
| 959 |
+
"learning_rate": 7.769600319330552e-05,
|
| 960 |
+
"loss": 0.1303,
|
| 961 |
+
"step": 3200
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.483885926269418,
|
| 965 |
+
"eval_loss": 0.16355698856626613,
|
| 966 |
+
"eval_runtime": 60.1946,
|
| 967 |
+
"eval_samples_per_second": 690.859,
|
| 968 |
+
"eval_steps_per_second": 0.681,
|
| 969 |
+
"step": 3200
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.5070716438673777,
|
| 973 |
+
"grad_norm": 0.06401767581701279,
|
| 974 |
+
"learning_rate": 7.703204087277988e-05,
|
| 975 |
+
"loss": 0.1315,
|
| 976 |
+
"step": 3250
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.5070716438673777,
|
| 980 |
+
"eval_loss": 0.16215006705140952,
|
| 981 |
+
"eval_runtime": 59.7822,
|
| 982 |
+
"eval_samples_per_second": 695.625,
|
| 983 |
+
"eval_steps_per_second": 0.686,
|
| 984 |
+
"step": 3250
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.5302573614653374,
|
| 988 |
+
"grad_norm": 0.07916898280382156,
|
| 989 |
+
"learning_rate": 7.636127338052512e-05,
|
| 990 |
+
"loss": 0.1315,
|
| 991 |
+
"step": 3300
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 1.5302573614653374,
|
| 995 |
+
"eval_loss": 0.16288597734760557,
|
| 996 |
+
"eval_runtime": 59.2757,
|
| 997 |
+
"eval_samples_per_second": 701.57,
|
| 998 |
+
"eval_steps_per_second": 0.692,
|
| 999 |
+
"step": 3300
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.553443079063297,
|
| 1003 |
+
"grad_norm": 0.06549016386270523,
|
| 1004 |
+
"learning_rate": 7.568386957867033e-05,
|
| 1005 |
+
"loss": 0.1303,
|
| 1006 |
+
"step": 3350
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.553443079063297,
|
| 1010 |
+
"eval_loss": 0.16416664097655873,
|
| 1011 |
+
"eval_runtime": 59.84,
|
| 1012 |
+
"eval_samples_per_second": 694.953,
|
| 1013 |
+
"eval_steps_per_second": 0.685,
|
| 1014 |
+
"step": 3350
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 1.5766287966612567,
|
| 1018 |
+
"grad_norm": 0.0709395632147789,
|
| 1019 |
+
"learning_rate": 7.500000000000001e-05,
|
| 1020 |
+
"loss": 0.1309,
|
| 1021 |
+
"step": 3400
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.5766287966612567,
|
| 1025 |
+
"eval_loss": 0.16179194486424098,
|
| 1026 |
+
"eval_runtime": 59.8634,
|
| 1027 |
+
"eval_samples_per_second": 694.682,
|
| 1028 |
+
"eval_steps_per_second": 0.685,
|
| 1029 |
+
"step": 3400
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 1.5998145142592164,
|
| 1033 |
+
"grad_norm": 0.05671363323926926,
|
| 1034 |
+
"learning_rate": 7.430983680502344e-05,
|
| 1035 |
+
"loss": 0.1307,
|
| 1036 |
+
"step": 3450
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 1.5998145142592164,
|
| 1040 |
+
"eval_loss": 0.16309191886303373,
|
| 1041 |
+
"eval_runtime": 59.618,
|
| 1042 |
+
"eval_samples_per_second": 697.541,
|
| 1043 |
+
"eval_steps_per_second": 0.688,
|
| 1044 |
+
"step": 3450
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.623000231857176,
|
| 1048 |
+
"grad_norm": 0.04889162629842758,
|
| 1049 |
+
"learning_rate": 7.361355373863414e-05,
|
| 1050 |
+
"loss": 0.1314,
|
| 1051 |
+
"step": 3500
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.623000231857176,
|
| 1055 |
+
"eval_loss": 0.16290782983414598,
|
| 1056 |
+
"eval_runtime": 60.3904,
|
| 1057 |
+
"eval_samples_per_second": 688.619,
|
| 1058 |
+
"eval_steps_per_second": 0.679,
|
| 1059 |
+
"step": 3500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.6461859494551356,
|
| 1063 |
+
"grad_norm": 0.0970933735370636,
|
| 1064 |
+
"learning_rate": 7.291132608637052e-05,
|
| 1065 |
+
"loss": 0.1314,
|
| 1066 |
+
"step": 3550
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.6461859494551356,
|
| 1070 |
+
"eval_loss": 0.16278222993823557,
|
| 1071 |
+
"eval_runtime": 59.8666,
|
| 1072 |
+
"eval_samples_per_second": 694.644,
|
| 1073 |
+
"eval_steps_per_second": 0.685,
|
| 1074 |
+
"step": 3550
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.6693716670530954,
|
| 1078 |
+
"grad_norm": 0.056557025760412216,
|
| 1079 |
+
"learning_rate": 7.220333063028872e-05,
|
| 1080 |
+
"loss": 0.1312,
|
| 1081 |
+
"step": 3600
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.6693716670530954,
|
| 1085 |
+
"eval_loss": 0.16313205291311117,
|
| 1086 |
+
"eval_runtime": 60.0092,
|
| 1087 |
+
"eval_samples_per_second": 692.993,
|
| 1088 |
+
"eval_steps_per_second": 0.683,
|
| 1089 |
+
"step": 3600
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.6925573846510549,
|
| 1093 |
+
"grad_norm": 0.04870522394776344,
|
| 1094 |
+
"learning_rate": 7.148974560445859e-05,
|
| 1095 |
+
"loss": 0.1299,
|
| 1096 |
+
"step": 3650
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.6925573846510549,
|
| 1100 |
+
"eval_loss": 0.1617941082289122,
|
| 1101 |
+
"eval_runtime": 60.1721,
|
| 1102 |
+
"eval_samples_per_second": 691.117,
|
| 1103 |
+
"eval_steps_per_second": 0.681,
|
| 1104 |
+
"step": 3650
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.7157431022490146,
|
| 1108 |
+
"grad_norm": 0.0681833028793335,
|
| 1109 |
+
"learning_rate": 7.077075065009433e-05,
|
| 1110 |
+
"loss": 0.1304,
|
| 1111 |
+
"step": 3700
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 1.7157431022490146,
|
| 1115 |
+
"eval_loss": 0.16243406602519425,
|
| 1116 |
+
"eval_runtime": 59.3626,
|
| 1117 |
+
"eval_samples_per_second": 700.542,
|
| 1118 |
+
"eval_steps_per_second": 0.691,
|
| 1119 |
+
"step": 3700
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 1.7389288198469743,
|
| 1123 |
+
"grad_norm": 0.06506156921386719,
|
| 1124 |
+
"learning_rate": 7.004652677033068e-05,
|
| 1125 |
+
"loss": 0.1299,
|
| 1126 |
+
"step": 3750
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 1.7389288198469743,
|
| 1130 |
+
"eval_loss": 0.16324780134312317,
|
| 1131 |
+
"eval_runtime": 59.6022,
|
| 1132 |
+
"eval_samples_per_second": 697.726,
|
| 1133 |
+
"eval_steps_per_second": 0.688,
|
| 1134 |
+
"step": 3750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 1.7621145374449338,
|
| 1138 |
+
"grad_norm": 0.06188170611858368,
|
| 1139 |
+
"learning_rate": 6.931725628465643e-05,
|
| 1140 |
+
"loss": 0.1309,
|
| 1141 |
+
"step": 3800
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 1.7621145374449338,
|
| 1145 |
+
"eval_loss": 0.1623115342294882,
|
| 1146 |
+
"eval_runtime": 59.7694,
|
| 1147 |
+
"eval_samples_per_second": 695.774,
|
| 1148 |
+
"eval_steps_per_second": 0.686,
|
| 1149 |
+
"step": 3800
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 1.7853002550428936,
|
| 1153 |
+
"grad_norm": 0.05675831064581871,
|
| 1154 |
+
"learning_rate": 6.858312278301637e-05,
|
| 1155 |
+
"loss": 0.1303,
|
| 1156 |
+
"step": 3850
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 1.7853002550428936,
|
| 1160 |
+
"eval_loss": 0.1630547638293529,
|
| 1161 |
+
"eval_runtime": 59.779,
|
| 1162 |
+
"eval_samples_per_second": 695.662,
|
| 1163 |
+
"eval_steps_per_second": 0.686,
|
| 1164 |
+
"step": 3850
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.8084859726408533,
|
| 1168 |
+
"grad_norm": 0.04727062210440636,
|
| 1169 |
+
"learning_rate": 6.784431107959359e-05,
|
| 1170 |
+
"loss": 0.1312,
|
| 1171 |
+
"step": 3900
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.8084859726408533,
|
| 1175 |
+
"eval_loss": 0.1616409071893626,
|
| 1176 |
+
"eval_runtime": 59.6005,
|
| 1177 |
+
"eval_samples_per_second": 697.746,
|
| 1178 |
+
"eval_steps_per_second": 0.688,
|
| 1179 |
+
"step": 3900
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.8316716902388128,
|
| 1183 |
+
"grad_norm": 0.06378892064094543,
|
| 1184 |
+
"learning_rate": 6.710100716628344e-05,
|
| 1185 |
+
"loss": 0.1303,
|
| 1186 |
+
"step": 3950
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.8316716902388128,
|
| 1190 |
+
"eval_loss": 0.1622395658739077,
|
| 1191 |
+
"eval_runtime": 60.1499,
|
| 1192 |
+
"eval_samples_per_second": 691.373,
|
| 1193 |
+
"eval_steps_per_second": 0.682,
|
| 1194 |
+
"step": 3950
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 1.8548574078367726,
|
| 1198 |
+
"grad_norm": 0.05470576509833336,
|
| 1199 |
+
"learning_rate": 6.635339816587109e-05,
|
| 1200 |
+
"loss": 0.1308,
|
| 1201 |
+
"step": 4000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.8548574078367726,
|
| 1205 |
+
"eval_loss": 0.16317236170181762,
|
| 1206 |
+
"eval_runtime": 60.014,
|
| 1207 |
+
"eval_samples_per_second": 692.939,
|
| 1208 |
+
"eval_steps_per_second": 0.683,
|
| 1209 |
+
"step": 4000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 1.8780431254347323,
|
| 1213 |
+
"grad_norm": 0.053886763751506805,
|
| 1214 |
+
"learning_rate": 6.560167228492436e-05,
|
| 1215 |
+
"loss": 0.1297,
|
| 1216 |
+
"step": 4050
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 1.8780431254347323,
|
| 1220 |
+
"eval_loss": 0.16198886262197804,
|
| 1221 |
+
"eval_runtime": 60.8262,
|
| 1222 |
+
"eval_samples_per_second": 683.685,
|
| 1223 |
+
"eval_steps_per_second": 0.674,
|
| 1224 |
+
"step": 4050
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.9012288430326918,
|
| 1228 |
+
"grad_norm": 0.054583676159381866,
|
| 1229 |
+
"learning_rate": 6.484601876641375e-05,
|
| 1230 |
+
"loss": 0.1301,
|
| 1231 |
+
"step": 4100
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 1.9012288430326918,
|
| 1235 |
+
"eval_loss": 0.1616550050294764,
|
| 1236 |
+
"eval_runtime": 59.7779,
|
| 1237 |
+
"eval_samples_per_second": 695.675,
|
| 1238 |
+
"eval_steps_per_second": 0.686,
|
| 1239 |
+
"step": 4100
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 1.9244145606306515,
|
| 1243 |
+
"grad_norm": 0.071171335875988,
|
| 1244 |
+
"learning_rate": 6.408662784207149e-05,
|
| 1245 |
+
"loss": 0.131,
|
| 1246 |
+
"step": 4150
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 1.9244145606306515,
|
| 1250 |
+
"eval_loss": 0.15968682813566223,
|
| 1251 |
+
"eval_runtime": 60.227,
|
| 1252 |
+
"eval_samples_per_second": 690.487,
|
| 1253 |
+
"eval_steps_per_second": 0.681,
|
| 1254 |
+
"step": 4150
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 1.9476002782286113,
|
| 1258 |
+
"grad_norm": 0.05775531381368637,
|
| 1259 |
+
"learning_rate": 6.332369068450174e-05,
|
| 1260 |
+
"loss": 0.1296,
|
| 1261 |
+
"step": 4200
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 1.9476002782286113,
|
| 1265 |
+
"eval_loss": 0.16262199212265846,
|
| 1266 |
+
"eval_runtime": 60.3405,
|
| 1267 |
+
"eval_samples_per_second": 689.189,
|
| 1268 |
+
"eval_steps_per_second": 0.679,
|
| 1269 |
+
"step": 4200
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.9707859958265708,
|
| 1273 |
+
"grad_norm": 0.06425776332616806,
|
| 1274 |
+
"learning_rate": 6.255739935905396e-05,
|
| 1275 |
+
"loss": 0.1299,
|
| 1276 |
+
"step": 4250
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.9707859958265708,
|
| 1280 |
+
"eval_loss": 0.16324524366491053,
|
| 1281 |
+
"eval_runtime": 61.417,
|
| 1282 |
+
"eval_samples_per_second": 677.109,
|
| 1283 |
+
"eval_steps_per_second": 0.668,
|
| 1284 |
+
"step": 4250
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.9939717134245305,
|
| 1288 |
+
"grad_norm": 0.045762140303850174,
|
| 1289 |
+
"learning_rate": 6.178794677547137e-05,
|
| 1290 |
+
"loss": 0.1299,
|
| 1291 |
+
"step": 4300
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.9939717134245305,
|
| 1295 |
+
"eval_loss": 0.16053301797614244,
|
| 1296 |
+
"eval_runtime": 61.0801,
|
| 1297 |
+
"eval_samples_per_second": 680.844,
|
| 1298 |
+
"eval_steps_per_second": 0.671,
|
| 1299 |
+
"step": 4300
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.0171574310224902,
|
| 1303 |
+
"grad_norm": 0.07060451060533524,
|
| 1304 |
+
"learning_rate": 6.1015526639327035e-05,
|
| 1305 |
+
"loss": 0.1296,
|
| 1306 |
+
"step": 4350
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.0171574310224902,
|
| 1310 |
+
"eval_loss": 0.1620254674138633,
|
| 1311 |
+
"eval_runtime": 61.0829,
|
| 1312 |
+
"eval_samples_per_second": 680.812,
|
| 1313 |
+
"eval_steps_per_second": 0.671,
|
| 1314 |
+
"step": 4350
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.0403431486204497,
|
| 1318 |
+
"grad_norm": 0.059919316321611404,
|
| 1319 |
+
"learning_rate": 6.024033340325954e-05,
|
| 1320 |
+
"loss": 0.1302,
|
| 1321 |
+
"step": 4400
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 2.0403431486204497,
|
| 1325 |
+
"eval_loss": 0.16284223807997533,
|
| 1326 |
+
"eval_runtime": 61.5789,
|
| 1327 |
+
"eval_samples_per_second": 675.328,
|
| 1328 |
+
"eval_steps_per_second": 0.666,
|
| 1329 |
+
"step": 4400
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.0635288662184093,
|
| 1333 |
+
"grad_norm": 0.07983385026454926,
|
| 1334 |
+
"learning_rate": 5.946256221802051e-05,
|
| 1335 |
+
"loss": 0.13,
|
| 1336 |
+
"step": 4450
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 2.0635288662184093,
|
| 1340 |
+
"eval_loss": 0.16209282393788932,
|
| 1341 |
+
"eval_runtime": 61.6584,
|
| 1342 |
+
"eval_samples_per_second": 674.458,
|
| 1343 |
+
"eval_steps_per_second": 0.665,
|
| 1344 |
+
"step": 4450
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.086714583816369,
|
| 1348 |
+
"grad_norm": 0.07582173496484756,
|
| 1349 |
+
"learning_rate": 5.868240888334653e-05,
|
| 1350 |
+
"loss": 0.1296,
|
| 1351 |
+
"step": 4500
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 2.086714583816369,
|
| 1355 |
+
"eval_loss": 0.16158196377974565,
|
| 1356 |
+
"eval_runtime": 61.1826,
|
| 1357 |
+
"eval_samples_per_second": 679.703,
|
| 1358 |
+
"eval_steps_per_second": 0.67,
|
| 1359 |
+
"step": 4500
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.1099003014143287,
|
| 1363 |
+
"grad_norm": 0.06049995869398117,
|
| 1364 |
+
"learning_rate": 5.79000697986675e-05,
|
| 1365 |
+
"loss": 0.1298,
|
| 1366 |
+
"step": 4550
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 2.1099003014143287,
|
| 1370 |
+
"eval_loss": 0.16130609279963956,
|
| 1371 |
+
"eval_runtime": 61.0153,
|
| 1372 |
+
"eval_samples_per_second": 681.567,
|
| 1373 |
+
"eval_steps_per_second": 0.672,
|
| 1374 |
+
"step": 4550
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.1330860190122882,
|
| 1378 |
+
"grad_norm": 0.0440148264169693,
|
| 1379 |
+
"learning_rate": 5.7115741913664264e-05,
|
| 1380 |
+
"loss": 0.1299,
|
| 1381 |
+
"step": 4600
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 2.1330860190122882,
|
| 1385 |
+
"eval_loss": 0.16027799763638953,
|
| 1386 |
+
"eval_runtime": 61.1993,
|
| 1387 |
+
"eval_samples_per_second": 679.517,
|
| 1388 |
+
"eval_steps_per_second": 0.67,
|
| 1389 |
+
"step": 4600
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.156271736610248,
|
| 1393 |
+
"grad_norm": 0.05254065990447998,
|
| 1394 |
+
"learning_rate": 5.6329622678687463e-05,
|
| 1395 |
+
"loss": 0.1299,
|
| 1396 |
+
"step": 4650
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.156271736610248,
|
| 1400 |
+
"eval_loss": 0.16206274484291652,
|
| 1401 |
+
"eval_runtime": 61.4415,
|
| 1402 |
+
"eval_samples_per_second": 676.839,
|
| 1403 |
+
"eval_steps_per_second": 0.667,
|
| 1404 |
+
"step": 4650
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.1794574542082077,
|
| 1408 |
+
"grad_norm": 0.06294432282447815,
|
| 1409 |
+
"learning_rate": 5.5541909995050554e-05,
|
| 1410 |
+
"loss": 0.1306,
|
| 1411 |
+
"step": 4700
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.1794574542082077,
|
| 1415 |
+
"eval_loss": 0.16140170723024802,
|
| 1416 |
+
"eval_runtime": 60.8861,
|
| 1417 |
+
"eval_samples_per_second": 683.013,
|
| 1418 |
+
"eval_steps_per_second": 0.673,
|
| 1419 |
+
"step": 4700
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.202643171806167,
|
| 1423 |
+
"grad_norm": 0.06710942089557648,
|
| 1424 |
+
"learning_rate": 5.475280216520913e-05,
|
| 1425 |
+
"loss": 0.1303,
|
| 1426 |
+
"step": 4750
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 2.202643171806167,
|
| 1430 |
+
"eval_loss": 0.16245448075670843,
|
| 1431 |
+
"eval_runtime": 61.2839,
|
| 1432 |
+
"eval_samples_per_second": 678.58,
|
| 1433 |
+
"eval_steps_per_second": 0.669,
|
| 1434 |
+
"step": 4750
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.225828889404127,
|
| 1438 |
+
"grad_norm": 0.05298132076859474,
|
| 1439 |
+
"learning_rate": 5.396249784283942e-05,
|
| 1440 |
+
"loss": 0.13,
|
| 1441 |
+
"step": 4800
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 2.225828889404127,
|
| 1445 |
+
"eval_loss": 0.1623738898660767,
|
| 1446 |
+
"eval_runtime": 61.1531,
|
| 1447 |
+
"eval_samples_per_second": 680.031,
|
| 1448 |
+
"eval_steps_per_second": 0.67,
|
| 1449 |
+
"step": 4800
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.2490146070020867,
|
| 1453 |
+
"grad_norm": 0.04066763445734978,
|
| 1454 |
+
"learning_rate": 5.317119598282823e-05,
|
| 1455 |
+
"loss": 0.1295,
|
| 1456 |
+
"step": 4850
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 2.2490146070020867,
|
| 1460 |
+
"eval_loss": 0.1627438727811327,
|
| 1461 |
+
"eval_runtime": 61.0414,
|
| 1462 |
+
"eval_samples_per_second": 681.275,
|
| 1463 |
+
"eval_steps_per_second": 0.672,
|
| 1464 |
+
"step": 4850
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 2.272200324600046,
|
| 1468 |
+
"grad_norm": 0.061821240931749344,
|
| 1469 |
+
"learning_rate": 5.2379095791187124e-05,
|
| 1470 |
+
"loss": 0.1299,
|
| 1471 |
+
"step": 4900
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 2.272200324600046,
|
| 1475 |
+
"eval_loss": 0.16086717177928397,
|
| 1476 |
+
"eval_runtime": 60.7945,
|
| 1477 |
+
"eval_samples_per_second": 684.042,
|
| 1478 |
+
"eval_steps_per_second": 0.674,
|
| 1479 |
+
"step": 4900
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 2.295386042198006,
|
| 1483 |
+
"grad_norm": 0.08038394153118134,
|
| 1484 |
+
"learning_rate": 5.158639667490339e-05,
|
| 1485 |
+
"loss": 0.13,
|
| 1486 |
+
"step": 4950
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.295386042198006,
|
| 1490 |
+
"eval_loss": 0.16221664317086187,
|
| 1491 |
+
"eval_runtime": 61.6742,
|
| 1492 |
+
"eval_samples_per_second": 674.285,
|
| 1493 |
+
"eval_steps_per_second": 0.665,
|
| 1494 |
+
"step": 4950
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.3185717597959656,
|
| 1498 |
+
"grad_norm": 0.0556926503777504,
|
| 1499 |
+
"learning_rate": 5.0793298191740404e-05,
|
| 1500 |
+
"loss": 0.1311,
|
| 1501 |
+
"step": 5000
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 2.3185717597959656,
|
| 1505 |
+
"eval_loss": 0.16015339844546791,
|
| 1506 |
+
"eval_runtime": 61.3657,
|
| 1507 |
+
"eval_samples_per_second": 677.675,
|
| 1508 |
+
"eval_steps_per_second": 0.668,
|
| 1509 |
+
"step": 5000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 2.3417574773939256,
|
| 1513 |
+
"grad_norm": 0.06645477563142776,
|
| 1514 |
+
"learning_rate": 5e-05,
|
| 1515 |
+
"loss": 0.1284,
|
| 1516 |
+
"step": 5050
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.3417574773939256,
|
| 1520 |
+
"eval_loss": 0.16160674186313023,
|
| 1521 |
+
"eval_runtime": 61.4737,
|
| 1522 |
+
"eval_samples_per_second": 676.484,
|
| 1523 |
+
"eval_steps_per_second": 0.667,
|
| 1524 |
+
"step": 5050
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 2.364943194991885,
|
| 1528 |
+
"grad_norm": 0.05365500971674919,
|
| 1529 |
+
"learning_rate": 4.92067018082596e-05,
|
| 1530 |
+
"loss": 0.13,
|
| 1531 |
+
"step": 5100
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 2.364943194991885,
|
| 1535 |
+
"eval_loss": 0.16016484459556096,
|
| 1536 |
+
"eval_runtime": 61.4058,
|
| 1537 |
+
"eval_samples_per_second": 677.232,
|
| 1538 |
+
"eval_steps_per_second": 0.668,
|
| 1539 |
+
"step": 5100
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 2.3881289125898446,
|
| 1543 |
+
"grad_norm": 0.0499204620718956,
|
| 1544 |
+
"learning_rate": 4.841360332509663e-05,
|
| 1545 |
+
"loss": 0.129,
|
| 1546 |
+
"step": 5150
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 2.3881289125898446,
|
| 1550 |
+
"eval_loss": 0.16054727464378063,
|
| 1551 |
+
"eval_runtime": 61.1539,
|
| 1552 |
+
"eval_samples_per_second": 680.023,
|
| 1553 |
+
"eval_steps_per_second": 0.67,
|
| 1554 |
+
"step": 5150
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 2.411314630187804,
|
| 1558 |
+
"grad_norm": 0.07284457236528397,
|
| 1559 |
+
"learning_rate": 4.762090420881289e-05,
|
| 1560 |
+
"loss": 0.129,
|
| 1561 |
+
"step": 5200
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 2.411314630187804,
|
| 1565 |
+
"eval_loss": 0.16057287778830004,
|
| 1566 |
+
"eval_runtime": 60.5785,
|
| 1567 |
+
"eval_samples_per_second": 686.481,
|
| 1568 |
+
"eval_steps_per_second": 0.677,
|
| 1569 |
+
"step": 5200
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 2.434500347785764,
|
| 1573 |
+
"grad_norm": 0.06511891633272171,
|
| 1574 |
+
"learning_rate": 4.6828804017171776e-05,
|
| 1575 |
+
"loss": 0.1297,
|
| 1576 |
+
"step": 5250
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 2.434500347785764,
|
| 1580 |
+
"eval_loss": 0.16202011190836896,
|
| 1581 |
+
"eval_runtime": 61.4053,
|
| 1582 |
+
"eval_samples_per_second": 677.238,
|
| 1583 |
+
"eval_steps_per_second": 0.668,
|
| 1584 |
+
"step": 5250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.4576860653837236,
|
| 1588 |
+
"grad_norm": 0.05936937406659126,
|
| 1589 |
+
"learning_rate": 4.603750215716057e-05,
|
| 1590 |
+
"loss": 0.1293,
|
| 1591 |
+
"step": 5300
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 2.4576860653837236,
|
| 1595 |
+
"eval_loss": 0.16067086041480225,
|
| 1596 |
+
"eval_runtime": 60.4469,
|
| 1597 |
+
"eval_samples_per_second": 687.976,
|
| 1598 |
+
"eval_steps_per_second": 0.678,
|
| 1599 |
+
"step": 5300
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 2.480871782981683,
|
| 1603 |
+
"grad_norm": 0.039836496114730835,
|
| 1604 |
+
"learning_rate": 4.5247197834790876e-05,
|
| 1605 |
+
"loss": 0.1288,
|
| 1606 |
+
"step": 5350
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 2.480871782981683,
|
| 1610 |
+
"eval_loss": 0.1614640227625451,
|
| 1611 |
+
"eval_runtime": 60.9513,
|
| 1612 |
+
"eval_samples_per_second": 682.283,
|
| 1613 |
+
"eval_steps_per_second": 0.673,
|
| 1614 |
+
"step": 5350
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 2.504057500579643,
|
| 1618 |
+
"grad_norm": 0.04305760934948921,
|
| 1619 |
+
"learning_rate": 4.445809000494946e-05,
|
| 1620 |
+
"loss": 0.1294,
|
| 1621 |
+
"step": 5400
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 2.504057500579643,
|
| 1625 |
+
"eval_loss": 0.16139181990447046,
|
| 1626 |
+
"eval_runtime": 60.6766,
|
| 1627 |
+
"eval_samples_per_second": 685.371,
|
| 1628 |
+
"eval_steps_per_second": 0.676,
|
| 1629 |
+
"step": 5400
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 2.5272432181776026,
|
| 1633 |
+
"grad_norm": 0.06780368089675903,
|
| 1634 |
+
"learning_rate": 4.3670377321312535e-05,
|
| 1635 |
+
"loss": 0.1285,
|
| 1636 |
+
"step": 5450
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 2.5272432181776026,
|
| 1640 |
+
"eval_loss": 0.1619736397134425,
|
| 1641 |
+
"eval_runtime": 60.7281,
|
| 1642 |
+
"eval_samples_per_second": 684.79,
|
| 1643 |
+
"eval_steps_per_second": 0.675,
|
| 1644 |
+
"step": 5450
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 2.550428935775562,
|
| 1648 |
+
"grad_norm": 0.052273835986852646,
|
| 1649 |
+
"learning_rate": 4.288425808633575e-05,
|
| 1650 |
+
"loss": 0.1303,
|
| 1651 |
+
"step": 5500
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 2.550428935775562,
|
| 1655 |
+
"eval_loss": 0.16178818674979198,
|
| 1656 |
+
"eval_runtime": 60.8875,
|
| 1657 |
+
"eval_samples_per_second": 682.997,
|
| 1658 |
+
"eval_steps_per_second": 0.673,
|
| 1659 |
+
"step": 5500
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 2.573614653373522,
|
| 1663 |
+
"grad_norm": 0.045574627816677094,
|
| 1664 |
+
"learning_rate": 4.20999302013325e-05,
|
| 1665 |
+
"loss": 0.1291,
|
| 1666 |
+
"step": 5550
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 2.573614653373522,
|
| 1670 |
+
"eval_loss": 0.16034006952877458,
|
| 1671 |
+
"eval_runtime": 60.7378,
|
| 1672 |
+
"eval_samples_per_second": 684.681,
|
| 1673 |
+
"eval_steps_per_second": 0.675,
|
| 1674 |
+
"step": 5550
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 2.5968003709714815,
|
| 1678 |
+
"grad_norm": 0.044092051684856415,
|
| 1679 |
+
"learning_rate": 4.131759111665349e-05,
|
| 1680 |
+
"loss": 0.1298,
|
| 1681 |
+
"step": 5600
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 2.5968003709714815,
|
| 1685 |
+
"eval_loss": 0.16090484909780667,
|
| 1686 |
+
"eval_runtime": 60.4675,
|
| 1687 |
+
"eval_samples_per_second": 687.741,
|
| 1688 |
+
"eval_steps_per_second": 0.678,
|
| 1689 |
+
"step": 5600
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 2.6199860885694415,
|
| 1693 |
+
"grad_norm": 0.05473971739411354,
|
| 1694 |
+
"learning_rate": 4.0537437781979506e-05,
|
| 1695 |
+
"loss": 0.1288,
|
| 1696 |
+
"step": 5650
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 2.6199860885694415,
|
| 1700 |
+
"eval_loss": 0.1604315377337276,
|
| 1701 |
+
"eval_runtime": 62.8239,
|
| 1702 |
+
"eval_samples_per_second": 661.946,
|
| 1703 |
+
"eval_steps_per_second": 0.653,
|
| 1704 |
+
"step": 5650
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.643171806167401,
|
| 1708 |
+
"grad_norm": 0.07100555300712585,
|
| 1709 |
+
"learning_rate": 3.9759666596740476e-05,
|
| 1710 |
+
"loss": 0.129,
|
| 1711 |
+
"step": 5700
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.643171806167401,
|
| 1715 |
+
"eval_loss": 0.15997494100305837,
|
| 1716 |
+
"eval_runtime": 61.3008,
|
| 1717 |
+
"eval_samples_per_second": 678.392,
|
| 1718 |
+
"eval_steps_per_second": 0.669,
|
| 1719 |
+
"step": 5700
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 2.6663575237653605,
|
| 1723 |
+
"grad_norm": 0.04020215570926666,
|
| 1724 |
+
"learning_rate": 3.898447336067297e-05,
|
| 1725 |
+
"loss": 0.1291,
|
| 1726 |
+
"step": 5750
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 2.6663575237653605,
|
| 1730 |
+
"eval_loss": 0.1596748490832133,
|
| 1731 |
+
"eval_runtime": 60.6148,
|
| 1732 |
+
"eval_samples_per_second": 686.07,
|
| 1733 |
+
"eval_steps_per_second": 0.676,
|
| 1734 |
+
"step": 5750
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 2.68954324136332,
|
| 1738 |
+
"grad_norm": 0.05526584014296532,
|
| 1739 |
+
"learning_rate": 3.821205322452863e-05,
|
| 1740 |
+
"loss": 0.1291,
|
| 1741 |
+
"step": 5800
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 2.68954324136332,
|
| 1745 |
+
"eval_loss": 0.16091962633426782,
|
| 1746 |
+
"eval_runtime": 60.1717,
|
| 1747 |
+
"eval_samples_per_second": 691.122,
|
| 1748 |
+
"eval_steps_per_second": 0.681,
|
| 1749 |
+
"step": 5800
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 2.71272895896128,
|
| 1753 |
+
"grad_norm": 0.052167922258377075,
|
| 1754 |
+
"learning_rate": 3.744260064094604e-05,
|
| 1755 |
+
"loss": 0.129,
|
| 1756 |
+
"step": 5850
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 2.71272895896128,
|
| 1760 |
+
"eval_loss": 0.16112806362615253,
|
| 1761 |
+
"eval_runtime": 60.1273,
|
| 1762 |
+
"eval_samples_per_second": 691.633,
|
| 1763 |
+
"eval_steps_per_second": 0.682,
|
| 1764 |
+
"step": 5850
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 2.7359146765592395,
|
| 1768 |
+
"grad_norm": 0.054320793598890305,
|
| 1769 |
+
"learning_rate": 3.6676309315498256e-05,
|
| 1770 |
+
"loss": 0.13,
|
| 1771 |
+
"step": 5900
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 2.7359146765592395,
|
| 1775 |
+
"eval_loss": 0.15996250695505343,
|
| 1776 |
+
"eval_runtime": 60.655,
|
| 1777 |
+
"eval_samples_per_second": 685.616,
|
| 1778 |
+
"eval_steps_per_second": 0.676,
|
| 1779 |
+
"step": 5900
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 2.7591003941571994,
|
| 1783 |
+
"grad_norm": 0.05470626428723335,
|
| 1784 |
+
"learning_rate": 3.591337215792852e-05,
|
| 1785 |
+
"loss": 0.1296,
|
| 1786 |
+
"step": 5950
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 2.7591003941571994,
|
| 1790 |
+
"eval_loss": 0.16025288890609335,
|
| 1791 |
+
"eval_runtime": 60.826,
|
| 1792 |
+
"eval_samples_per_second": 683.688,
|
| 1793 |
+
"eval_steps_per_second": 0.674,
|
| 1794 |
+
"step": 5950
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 2.782286111755159,
|
| 1798 |
+
"grad_norm": 0.04805810749530792,
|
| 1799 |
+
"learning_rate": 3.515398123358627e-05,
|
| 1800 |
+
"loss": 0.1294,
|
| 1801 |
+
"step": 6000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 2.782286111755159,
|
| 1805 |
+
"eval_loss": 0.15918263724182835,
|
| 1806 |
+
"eval_runtime": 60.2321,
|
| 1807 |
+
"eval_samples_per_second": 690.429,
|
| 1808 |
+
"eval_steps_per_second": 0.681,
|
| 1809 |
+
"step": 6000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.8054718293531185,
|
| 1813 |
+
"grad_norm": 0.04185302183032036,
|
| 1814 |
+
"learning_rate": 3.439832771507565e-05,
|
| 1815 |
+
"loss": 0.1283,
|
| 1816 |
+
"step": 6050
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.8054718293531185,
|
| 1820 |
+
"eval_loss": 0.16179385240233157,
|
| 1821 |
+
"eval_runtime": 60.9176,
|
| 1822 |
+
"eval_samples_per_second": 682.66,
|
| 1823 |
+
"eval_steps_per_second": 0.673,
|
| 1824 |
+
"step": 6050
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 2.828657546951078,
|
| 1828 |
+
"grad_norm": 0.04609336704015732,
|
| 1829 |
+
"learning_rate": 3.364660183412892e-05,
|
| 1830 |
+
"loss": 0.1292,
|
| 1831 |
+
"step": 6100
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 2.828657546951078,
|
| 1835 |
+
"eval_loss": 0.1611929898635588,
|
| 1836 |
+
"eval_runtime": 60.5916,
|
| 1837 |
+
"eval_samples_per_second": 686.333,
|
| 1838 |
+
"eval_steps_per_second": 0.677,
|
| 1839 |
+
"step": 6100
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 2.851843264549038,
|
| 1843 |
+
"grad_norm": 0.05404876172542572,
|
| 1844 |
+
"learning_rate": 3.289899283371657e-05,
|
| 1845 |
+
"loss": 0.128,
|
| 1846 |
+
"step": 6150
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 2.851843264549038,
|
| 1850 |
+
"eval_loss": 0.16039360794951976,
|
| 1851 |
+
"eval_runtime": 60.5961,
|
| 1852 |
+
"eval_samples_per_second": 686.282,
|
| 1853 |
+
"eval_steps_per_second": 0.677,
|
| 1854 |
+
"step": 6150
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 2.8750289821469974,
|
| 1858 |
+
"grad_norm": 0.06787659227848053,
|
| 1859 |
+
"learning_rate": 3.215568892040641e-05,
|
| 1860 |
+
"loss": 0.1288,
|
| 1861 |
+
"step": 6200
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 2.8750289821469974,
|
| 1865 |
+
"eval_loss": 0.16113480515361805,
|
| 1866 |
+
"eval_runtime": 60.2775,
|
| 1867 |
+
"eval_samples_per_second": 689.909,
|
| 1868 |
+
"eval_steps_per_second": 0.68,
|
| 1869 |
+
"step": 6200
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 2.8982146997449574,
|
| 1873 |
+
"grad_norm": 0.06937435269355774,
|
| 1874 |
+
"learning_rate": 3.141687721698363e-05,
|
| 1875 |
+
"loss": 0.1283,
|
| 1876 |
+
"step": 6250
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 2.8982146997449574,
|
| 1880 |
+
"eval_loss": 0.16087572214972407,
|
| 1881 |
+
"eval_runtime": 60.6789,
|
| 1882 |
+
"eval_samples_per_second": 685.345,
|
| 1883 |
+
"eval_steps_per_second": 0.676,
|
| 1884 |
+
"step": 6250
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 2.921400417342917,
|
| 1888 |
+
"grad_norm": 0.08074232190847397,
|
| 1889 |
+
"learning_rate": 3.0682743715343564e-05,
|
| 1890 |
+
"loss": 0.1292,
|
| 1891 |
+
"step": 6300
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 2.921400417342917,
|
| 1895 |
+
"eval_loss": 0.16049740787316144,
|
| 1896 |
+
"eval_runtime": 60.3194,
|
| 1897 |
+
"eval_samples_per_second": 689.43,
|
| 1898 |
+
"eval_steps_per_second": 0.68,
|
| 1899 |
+
"step": 6300
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 2.9445861349408764,
|
| 1903 |
+
"grad_norm": 0.03976515680551529,
|
| 1904 |
+
"learning_rate": 2.9953473229669328e-05,
|
| 1905 |
+
"loss": 0.1302,
|
| 1906 |
+
"step": 6350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.9445861349408764,
|
| 1910 |
+
"eval_loss": 0.16023700059761273,
|
| 1911 |
+
"eval_runtime": 60.8537,
|
| 1912 |
+
"eval_samples_per_second": 683.377,
|
| 1913 |
+
"eval_steps_per_second": 0.674,
|
| 1914 |
+
"step": 6350
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.967771852538836,
|
| 1918 |
+
"grad_norm": 0.05303976684808731,
|
| 1919 |
+
"learning_rate": 2.9229249349905684e-05,
|
| 1920 |
+
"loss": 0.1285,
|
| 1921 |
+
"step": 6400
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.967771852538836,
|
| 1925 |
+
"eval_loss": 0.1601465398516622,
|
| 1926 |
+
"eval_runtime": 60.6472,
|
| 1927 |
+
"eval_samples_per_second": 685.703,
|
| 1928 |
+
"eval_steps_per_second": 0.676,
|
| 1929 |
+
"step": 6400
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 2.990957570136796,
|
| 1933 |
+
"grad_norm": 0.0519745759665966,
|
| 1934 |
+
"learning_rate": 2.851025439554142e-05,
|
| 1935 |
+
"loss": 0.1286,
|
| 1936 |
+
"step": 6450
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 2.990957570136796,
|
| 1940 |
+
"eval_loss": 0.16085429229133483,
|
| 1941 |
+
"eval_runtime": 60.2507,
|
| 1942 |
+
"eval_samples_per_second": 690.216,
|
| 1943 |
+
"eval_steps_per_second": 0.68,
|
| 1944 |
+
"step": 6450
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 3.0141432877347554,
|
| 1948 |
+
"grad_norm": 0.050518251955509186,
|
| 1949 |
+
"learning_rate": 2.7796669369711294e-05,
|
| 1950 |
+
"loss": 0.1301,
|
| 1951 |
+
"step": 6500
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 3.0141432877347554,
|
| 1955 |
+
"eval_loss": 0.16015394660421692,
|
| 1956 |
+
"eval_runtime": 60.5015,
|
| 1957 |
+
"eval_samples_per_second": 687.355,
|
| 1958 |
+
"eval_steps_per_second": 0.678,
|
| 1959 |
+
"step": 6500
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 3.037329005332715,
|
| 1963 |
+
"grad_norm": 0.04253960773348808,
|
| 1964 |
+
"learning_rate": 2.708867391362948e-05,
|
| 1965 |
+
"loss": 0.1296,
|
| 1966 |
+
"step": 6550
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 3.037329005332715,
|
| 1970 |
+
"eval_loss": 0.1597283595131218,
|
| 1971 |
+
"eval_runtime": 60.13,
|
| 1972 |
+
"eval_samples_per_second": 691.601,
|
| 1973 |
+
"eval_steps_per_second": 0.682,
|
| 1974 |
+
"step": 6550
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 3.060514722930675,
|
| 1978 |
+
"grad_norm": 0.06899340450763702,
|
| 1979 |
+
"learning_rate": 2.638644626136587e-05,
|
| 1980 |
+
"loss": 0.1291,
|
| 1981 |
+
"step": 6600
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 3.060514722930675,
|
| 1985 |
+
"eval_loss": 0.1604277250117246,
|
| 1986 |
+
"eval_runtime": 60.4618,
|
| 1987 |
+
"eval_samples_per_second": 687.806,
|
| 1988 |
+
"eval_steps_per_second": 0.678,
|
| 1989 |
+
"step": 6600
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 3.0837004405286343,
|
| 1993 |
+
"grad_norm": 0.06556117534637451,
|
| 1994 |
+
"learning_rate": 2.5690163194976575e-05,
|
| 1995 |
+
"loss": 0.1288,
|
| 1996 |
+
"step": 6650
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 3.0837004405286343,
|
| 2000 |
+
"eval_loss": 0.15953636330193482,
|
| 2001 |
+
"eval_runtime": 60.2757,
|
| 2002 |
+
"eval_samples_per_second": 689.93,
|
| 2003 |
+
"eval_steps_per_second": 0.68,
|
| 2004 |
+
"step": 6650
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 3.106886158126594,
|
| 2008 |
+
"grad_norm": 0.03685734421014786,
|
| 2009 |
+
"learning_rate": 2.500000000000001e-05,
|
| 2010 |
+
"loss": 0.129,
|
| 2011 |
+
"step": 6700
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 3.106886158126594,
|
| 2015 |
+
"eval_loss": 0.159308270335797,
|
| 2016 |
+
"eval_runtime": 60.624,
|
| 2017 |
+
"eval_samples_per_second": 685.966,
|
| 2018 |
+
"eval_steps_per_second": 0.676,
|
| 2019 |
+
"step": 6700
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 3.130071875724554,
|
| 2023 |
+
"grad_norm": 0.0451020672917366,
|
| 2024 |
+
"learning_rate": 2.4316130421329697e-05,
|
| 2025 |
+
"loss": 0.1286,
|
| 2026 |
+
"step": 6750
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 3.130071875724554,
|
| 2030 |
+
"eval_loss": 0.15995884031774596,
|
| 2031 |
+
"eval_runtime": 60.3654,
|
| 2032 |
+
"eval_samples_per_second": 688.905,
|
| 2033 |
+
"eval_steps_per_second": 0.679,
|
| 2034 |
+
"step": 6750
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 3.1532575933225133,
|
| 2038 |
+
"grad_norm": 0.0495733842253685,
|
| 2039 |
+
"learning_rate": 2.363872661947488e-05,
|
| 2040 |
+
"loss": 0.1293,
|
| 2041 |
+
"step": 6800
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 3.1532575933225133,
|
| 2045 |
+
"eval_loss": 0.15987331824692497,
|
| 2046 |
+
"eval_runtime": 60.4636,
|
| 2047 |
+
"eval_samples_per_second": 687.786,
|
| 2048 |
+
"eval_steps_per_second": 0.678,
|
| 2049 |
+
"step": 6800
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 3.176443310920473,
|
| 2053 |
+
"grad_norm": 0.05756652355194092,
|
| 2054 |
+
"learning_rate": 2.296795912722014e-05,
|
| 2055 |
+
"loss": 0.1289,
|
| 2056 |
+
"step": 6850
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 3.176443310920473,
|
| 2060 |
+
"eval_loss": 0.15986134614331013,
|
| 2061 |
+
"eval_runtime": 61.0063,
|
| 2062 |
+
"eval_samples_per_second": 681.667,
|
| 2063 |
+
"eval_steps_per_second": 0.672,
|
| 2064 |
+
"step": 6850
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 3.199629028518433,
|
| 2068 |
+
"grad_norm": 0.0467820018529892,
|
| 2069 |
+
"learning_rate": 2.2303996806694488e-05,
|
| 2070 |
+
"loss": 0.1295,
|
| 2071 |
+
"step": 6900
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 3.199629028518433,
|
| 2075 |
+
"eval_loss": 0.16011030076900337,
|
| 2076 |
+
"eval_runtime": 60.1041,
|
| 2077 |
+
"eval_samples_per_second": 691.9,
|
| 2078 |
+
"eval_steps_per_second": 0.682,
|
| 2079 |
+
"step": 6900
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 3.2228147461163923,
|
| 2083 |
+
"grad_norm": 0.04179982468485832,
|
| 2084 |
+
"learning_rate": 2.164700680686147e-05,
|
| 2085 |
+
"loss": 0.1287,
|
| 2086 |
+
"step": 6950
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 3.2228147461163923,
|
| 2090 |
+
"eval_loss": 0.15917751068552838,
|
| 2091 |
+
"eval_runtime": 60.5321,
|
| 2092 |
+
"eval_samples_per_second": 687.007,
|
| 2093 |
+
"eval_steps_per_second": 0.677,
|
| 2094 |
+
"step": 6950
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 3.246000463714352,
|
| 2098 |
+
"grad_norm": 0.053910572081804276,
|
| 2099 |
+
"learning_rate": 2.09971545214401e-05,
|
| 2100 |
+
"loss": 0.1286,
|
| 2101 |
+
"step": 7000
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 3.246000463714352,
|
| 2105 |
+
"eval_loss": 0.15998092838627764,
|
| 2106 |
+
"eval_runtime": 60.4067,
|
| 2107 |
+
"eval_samples_per_second": 688.434,
|
| 2108 |
+
"eval_steps_per_second": 0.679,
|
| 2109 |
+
"step": 7000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 3.2691861813123118,
|
| 2113 |
+
"grad_norm": 0.04404950886964798,
|
| 2114 |
+
"learning_rate": 2.0354603547267985e-05,
|
| 2115 |
+
"loss": 0.1283,
|
| 2116 |
+
"step": 7050
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 3.2691861813123118,
|
| 2120 |
+
"eval_loss": 0.1597617331551387,
|
| 2121 |
+
"eval_runtime": 60.4218,
|
| 2122 |
+
"eval_samples_per_second": 688.262,
|
| 2123 |
+
"eval_steps_per_second": 0.679,
|
| 2124 |
+
"step": 7050
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 3.2923718989102713,
|
| 2128 |
+
"grad_norm": 0.04763752967119217,
|
| 2129 |
+
"learning_rate": 1.9719515643116674e-05,
|
| 2130 |
+
"loss": 0.1288,
|
| 2131 |
+
"step": 7100
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 3.2923718989102713,
|
| 2135 |
+
"eval_loss": 0.16116006530852447,
|
| 2136 |
+
"eval_runtime": 60.2132,
|
| 2137 |
+
"eval_samples_per_second": 690.646,
|
| 2138 |
+
"eval_steps_per_second": 0.681,
|
| 2139 |
+
"step": 7100
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 3.3155576165082308,
|
| 2143 |
+
"grad_norm": 0.049567196518182755,
|
| 2144 |
+
"learning_rate": 1.9092050688969738e-05,
|
| 2145 |
+
"loss": 0.1298,
|
| 2146 |
+
"step": 7150
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 3.3155576165082308,
|
| 2150 |
+
"eval_loss": 0.15965543804361845,
|
| 2151 |
+
"eval_runtime": 60.3928,
|
| 2152 |
+
"eval_samples_per_second": 688.592,
|
| 2153 |
+
"eval_steps_per_second": 0.679,
|
| 2154 |
+
"step": 7150
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 3.3387433341061907,
|
| 2158 |
+
"grad_norm": 0.05488676205277443,
|
| 2159 |
+
"learning_rate": 1.847236664577389e-05,
|
| 2160 |
+
"loss": 0.1284,
|
| 2161 |
+
"step": 7200
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 3.3387433341061907,
|
| 2165 |
+
"eval_loss": 0.16050384662882064,
|
| 2166 |
+
"eval_runtime": 60.121,
|
| 2167 |
+
"eval_samples_per_second": 691.705,
|
| 2168 |
+
"eval_steps_per_second": 0.682,
|
| 2169 |
+
"step": 7200
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 3.3619290517041502,
|
| 2173 |
+
"grad_norm": 0.04124298691749573,
|
| 2174 |
+
"learning_rate": 1.7860619515673033e-05,
|
| 2175 |
+
"loss": 0.1289,
|
| 2176 |
+
"step": 7250
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 3.3619290517041502,
|
| 2180 |
+
"eval_loss": 0.16054145931691394,
|
| 2181 |
+
"eval_runtime": 60.2046,
|
| 2182 |
+
"eval_samples_per_second": 690.745,
|
| 2183 |
+
"eval_steps_per_second": 0.681,
|
| 2184 |
+
"step": 7250
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 3.3851147693021097,
|
| 2188 |
+
"grad_norm": 0.04400424286723137,
|
| 2189 |
+
"learning_rate": 1.725696330273575e-05,
|
| 2190 |
+
"loss": 0.1289,
|
| 2191 |
+
"step": 7300
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 3.3851147693021097,
|
| 2195 |
+
"eval_loss": 0.15999099129576416,
|
| 2196 |
+
"eval_runtime": 60.4869,
|
| 2197 |
+
"eval_samples_per_second": 687.52,
|
| 2198 |
+
"eval_steps_per_second": 0.678,
|
| 2199 |
+
"step": 7300
|
| 2200 |
+
},
|
| 2201 |
+
{
|
| 2202 |
+
"epoch": 3.4083004869000697,
|
| 2203 |
+
"grad_norm": 0.05488509312272072,
|
| 2204 |
+
"learning_rate": 1.6661549974185424e-05,
|
| 2205 |
+
"loss": 0.1285,
|
| 2206 |
+
"step": 7350
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 3.4083004869000697,
|
| 2210 |
+
"eval_loss": 0.16051823730892306,
|
| 2211 |
+
"eval_runtime": 60.1981,
|
| 2212 |
+
"eval_samples_per_second": 690.819,
|
| 2213 |
+
"eval_steps_per_second": 0.681,
|
| 2214 |
+
"step": 7350
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 3.431486204498029,
|
| 2218 |
+
"grad_norm": 0.06722457706928253,
|
| 2219 |
+
"learning_rate": 1.60745294221434e-05,
|
| 2220 |
+
"loss": 0.1286,
|
| 2221 |
+
"step": 7400
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 3.431486204498029,
|
| 2225 |
+
"eval_loss": 0.1610307768591294,
|
| 2226 |
+
"eval_runtime": 60.7755,
|
| 2227 |
+
"eval_samples_per_second": 684.256,
|
| 2228 |
+
"eval_steps_per_second": 0.675,
|
| 2229 |
+
"step": 7400
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 3.4546719220959887,
|
| 2233 |
+
"grad_norm": 0.04814394935965538,
|
| 2234 |
+
"learning_rate": 1.549604942589441e-05,
|
| 2235 |
+
"loss": 0.1278,
|
| 2236 |
+
"step": 7450
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 3.4546719220959887,
|
| 2240 |
+
"eval_loss": 0.1598065741965525,
|
| 2241 |
+
"eval_runtime": 59.9968,
|
| 2242 |
+
"eval_samples_per_second": 693.136,
|
| 2243 |
+
"eval_steps_per_second": 0.683,
|
| 2244 |
+
"step": 7450
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 3.4778576396939487,
|
| 2248 |
+
"grad_norm": 0.04934167116880417,
|
| 2249 |
+
"learning_rate": 1.4926255614683932e-05,
|
| 2250 |
+
"loss": 0.1274,
|
| 2251 |
+
"step": 7500
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 3.4778576396939487,
|
| 2255 |
+
"eval_loss": 0.15982454893723042,
|
| 2256 |
+
"eval_runtime": 60.201,
|
| 2257 |
+
"eval_samples_per_second": 690.786,
|
| 2258 |
+
"eval_steps_per_second": 0.681,
|
| 2259 |
+
"step": 7500
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 3.501043357291908,
|
| 2263 |
+
"grad_norm": 0.04529615864157677,
|
| 2264 |
+
"learning_rate": 1.4365291431056871e-05,
|
| 2265 |
+
"loss": 0.1297,
|
| 2266 |
+
"step": 7550
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 3.501043357291908,
|
| 2270 |
+
"eval_loss": 0.15986133524024926,
|
| 2271 |
+
"eval_runtime": 59.95,
|
| 2272 |
+
"eval_samples_per_second": 693.678,
|
| 2273 |
+
"eval_steps_per_second": 0.684,
|
| 2274 |
+
"step": 7550
|
| 2275 |
+
},
|
| 2276 |
+
{
|
| 2277 |
+
"epoch": 3.5242290748898677,
|
| 2278 |
+
"grad_norm": 0.0399620421230793,
|
| 2279 |
+
"learning_rate": 1.3813298094746491e-05,
|
| 2280 |
+
"loss": 0.1288,
|
| 2281 |
+
"step": 7600
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 3.5242290748898677,
|
| 2285 |
+
"eval_loss": 0.15905609221590689,
|
| 2286 |
+
"eval_runtime": 61.181,
|
| 2287 |
+
"eval_samples_per_second": 679.72,
|
| 2288 |
+
"eval_steps_per_second": 0.67,
|
| 2289 |
+
"step": 7600
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 3.5474147924878277,
|
| 2293 |
+
"grad_norm": 0.05973295867443085,
|
| 2294 |
+
"learning_rate": 1.327041456712334e-05,
|
| 2295 |
+
"loss": 0.1281,
|
| 2296 |
+
"step": 7650
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 3.5474147924878277,
|
| 2300 |
+
"eval_loss": 0.15981091550942805,
|
| 2301 |
+
"eval_runtime": 60.5605,
|
| 2302 |
+
"eval_samples_per_second": 686.685,
|
| 2303 |
+
"eval_steps_per_second": 0.677,
|
| 2304 |
+
"step": 7650
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 3.570600510085787,
|
| 2308 |
+
"grad_norm": 0.04896661266684532,
|
| 2309 |
+
"learning_rate": 1.2736777516212266e-05,
|
| 2310 |
+
"loss": 0.1288,
|
| 2311 |
+
"step": 7700
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"epoch": 3.570600510085787,
|
| 2315 |
+
"eval_loss": 0.1599924400443614,
|
| 2316 |
+
"eval_runtime": 60.486,
|
| 2317 |
+
"eval_samples_per_second": 687.531,
|
| 2318 |
+
"eval_steps_per_second": 0.678,
|
| 2319 |
+
"step": 7700
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 3.5937862276837467,
|
| 2323 |
+
"grad_norm": 0.07458525151014328,
|
| 2324 |
+
"learning_rate": 1.2212521282287092e-05,
|
| 2325 |
+
"loss": 0.128,
|
| 2326 |
+
"step": 7750
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 3.5937862276837467,
|
| 2330 |
+
"eval_loss": 0.15936126278835275,
|
| 2331 |
+
"eval_runtime": 60.9341,
|
| 2332 |
+
"eval_samples_per_second": 682.475,
|
| 2333 |
+
"eval_steps_per_second": 0.673,
|
| 2334 |
+
"step": 7750
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 3.6169719452817066,
|
| 2338 |
+
"grad_norm": 0.04200127348303795,
|
| 2339 |
+
"learning_rate": 1.1697777844051105e-05,
|
| 2340 |
+
"loss": 0.1287,
|
| 2341 |
+
"step": 7800
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 3.6169719452817066,
|
| 2345 |
+
"eval_loss": 0.1603394617678833,
|
| 2346 |
+
"eval_runtime": 60.5155,
|
| 2347 |
+
"eval_samples_per_second": 687.195,
|
| 2348 |
+
"eval_steps_per_second": 0.678,
|
| 2349 |
+
"step": 7800
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 3.640157662879666,
|
| 2353 |
+
"grad_norm": 0.06712640821933746,
|
| 2354 |
+
"learning_rate": 1.1192676785412154e-05,
|
| 2355 |
+
"loss": 0.1291,
|
| 2356 |
+
"step": 7850
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 3.640157662879666,
|
| 2360 |
+
"eval_loss": 0.15920067938345067,
|
| 2361 |
+
"eval_runtime": 60.0225,
|
| 2362 |
+
"eval_samples_per_second": 692.84,
|
| 2363 |
+
"eval_steps_per_second": 0.683,
|
| 2364 |
+
"step": 7850
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"epoch": 3.6633433804776256,
|
| 2368 |
+
"grad_norm": 0.049462996423244476,
|
| 2369 |
+
"learning_rate": 1.0697345262860636e-05,
|
| 2370 |
+
"loss": 0.1287,
|
| 2371 |
+
"step": 7900
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 3.6633433804776256,
|
| 2375 |
+
"eval_loss": 0.15964593569874527,
|
| 2376 |
+
"eval_runtime": 60.1965,
|
| 2377 |
+
"eval_samples_per_second": 690.837,
|
| 2378 |
+
"eval_steps_per_second": 0.681,
|
| 2379 |
+
"step": 7900
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 3.6865290980755856,
|
| 2383 |
+
"grad_norm": 0.05148932337760925,
|
| 2384 |
+
"learning_rate": 1.021190797345839e-05,
|
| 2385 |
+
"loss": 0.1283,
|
| 2386 |
+
"step": 7950
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 3.6865290980755856,
|
| 2390 |
+
"eval_loss": 0.15903354419354673,
|
| 2391 |
+
"eval_runtime": 60.0507,
|
| 2392 |
+
"eval_samples_per_second": 692.515,
|
| 2393 |
+
"eval_steps_per_second": 0.683,
|
| 2394 |
+
"step": 7950
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 3.709714815673545,
|
| 2398 |
+
"grad_norm": 0.05164024233818054,
|
| 2399 |
+
"learning_rate": 9.73648712344707e-06,
|
| 2400 |
+
"loss": 0.128,
|
| 2401 |
+
"step": 8000
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 3.709714815673545,
|
| 2405 |
+
"eval_loss": 0.15835035051131605,
|
| 2406 |
+
"eval_runtime": 60.5739,
|
| 2407 |
+
"eval_samples_per_second": 686.533,
|
| 2408 |
+
"eval_steps_per_second": 0.677,
|
| 2409 |
+
"step": 8000
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 3.7329005332715046,
|
| 2413 |
+
"grad_norm": 0.04926716163754463,
|
| 2414 |
+
"learning_rate": 9.271202397483215e-06,
|
| 2415 |
+
"loss": 0.1276,
|
| 2416 |
+
"step": 8050
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 3.7329005332715046,
|
| 2420 |
+
"eval_loss": 0.160225615529793,
|
| 2421 |
+
"eval_runtime": 60.4555,
|
| 2422 |
+
"eval_samples_per_second": 687.878,
|
| 2423 |
+
"eval_steps_per_second": 0.678,
|
| 2424 |
+
"step": 8050
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 3.7560862508694646,
|
| 2428 |
+
"grad_norm": 0.04355842247605324,
|
| 2429 |
+
"learning_rate": 8.816170928508365e-06,
|
| 2430 |
+
"loss": 0.1287,
|
| 2431 |
+
"step": 8100
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 3.7560862508694646,
|
| 2435 |
+
"eval_loss": 0.1601867779420742,
|
| 2436 |
+
"eval_runtime": 60.7386,
|
| 2437 |
+
"eval_samples_per_second": 684.672,
|
| 2438 |
+
"eval_steps_per_second": 0.675,
|
| 2439 |
+
"step": 8100
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 3.779271968467424,
|
| 2443 |
+
"grad_norm": 0.039105553179979324,
|
| 2444 |
+
"learning_rate": 8.371507268261437e-06,
|
| 2445 |
+
"loss": 0.1306,
|
| 2446 |
+
"step": 8150
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 3.779271968467424,
|
| 2450 |
+
"eval_loss": 0.15946348937187382,
|
| 2451 |
+
"eval_runtime": 60.9253,
|
| 2452 |
+
"eval_samples_per_second": 682.574,
|
| 2453 |
+
"eval_steps_per_second": 0.673,
|
| 2454 |
+
"step": 8150
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 3.8024576860653836,
|
| 2458 |
+
"grad_norm": 0.04452899843454361,
|
| 2459 |
+
"learning_rate": 7.937323358440935e-06,
|
| 2460 |
+
"loss": 0.1286,
|
| 2461 |
+
"step": 8200
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 3.8024576860653836,
|
| 2465 |
+
"eval_loss": 0.15871429728364056,
|
| 2466 |
+
"eval_runtime": 60.2776,
|
| 2467 |
+
"eval_samples_per_second": 689.908,
|
| 2468 |
+
"eval_steps_per_second": 0.68,
|
| 2469 |
+
"step": 8200
|
| 2470 |
+
},
|
| 2471 |
+
{
|
| 2472 |
+
"epoch": 3.8256434036633435,
|
| 2473 |
+
"grad_norm": 0.043075498193502426,
|
| 2474 |
+
"learning_rate": 7.513728502524286e-06,
|
| 2475 |
+
"loss": 0.1292,
|
| 2476 |
+
"step": 8250
|
| 2477 |
+
},
|
| 2478 |
+
{
|
| 2479 |
+
"epoch": 3.8256434036633435,
|
| 2480 |
+
"eval_loss": 0.1592580359542711,
|
| 2481 |
+
"eval_runtime": 60.7244,
|
| 2482 |
+
"eval_samples_per_second": 684.832,
|
| 2483 |
+
"eval_steps_per_second": 0.675,
|
| 2484 |
+
"step": 8250
|
| 2485 |
+
},
|
| 2486 |
+
{
|
| 2487 |
+
"epoch": 3.848829121261303,
|
| 2488 |
+
"grad_norm": 0.05848800390958786,
|
| 2489 |
+
"learning_rate": 7.100829338251147e-06,
|
| 2490 |
+
"loss": 0.1275,
|
| 2491 |
+
"step": 8300
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 3.848829121261303,
|
| 2495 |
+
"eval_loss": 0.15895083163665807,
|
| 2496 |
+
"eval_runtime": 60.3677,
|
| 2497 |
+
"eval_samples_per_second": 688.878,
|
| 2498 |
+
"eval_steps_per_second": 0.679,
|
| 2499 |
+
"step": 8300
|
| 2500 |
+
},
|
| 2501 |
+
{
|
| 2502 |
+
"epoch": 3.8720148388592626,
|
| 2503 |
+
"grad_norm": 0.04980336129665375,
|
| 2504 |
+
"learning_rate": 6.698729810778065e-06,
|
| 2505 |
+
"loss": 0.1277,
|
| 2506 |
+
"step": 8350
|
| 2507 |
+
},
|
| 2508 |
+
{
|
| 2509 |
+
"epoch": 3.8720148388592626,
|
| 2510 |
+
"eval_loss": 0.16002303550437,
|
| 2511 |
+
"eval_runtime": 60.2742,
|
| 2512 |
+
"eval_samples_per_second": 689.947,
|
| 2513 |
+
"eval_steps_per_second": 0.68,
|
| 2514 |
+
"step": 8350
|
| 2515 |
+
},
|
| 2516 |
+
{
|
| 2517 |
+
"epoch": 3.8952005564572225,
|
| 2518 |
+
"grad_norm": 0.057385146617889404,
|
| 2519 |
+
"learning_rate": 6.3075311465107535e-06,
|
| 2520 |
+
"loss": 0.129,
|
| 2521 |
+
"step": 8400
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 3.8952005564572225,
|
| 2525 |
+
"eval_loss": 0.1601535826416112,
|
| 2526 |
+
"eval_runtime": 60.4053,
|
| 2527 |
+
"eval_samples_per_second": 688.45,
|
| 2528 |
+
"eval_steps_per_second": 0.679,
|
| 2529 |
+
"step": 8400
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 3.918386274055182,
|
| 2533 |
+
"grad_norm": 0.045788682997226715,
|
| 2534 |
+
"learning_rate": 5.927331827620903e-06,
|
| 2535 |
+
"loss": 0.1286,
|
| 2536 |
+
"step": 8450
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 3.918386274055182,
|
| 2540 |
+
"eval_loss": 0.15926720973175468,
|
| 2541 |
+
"eval_runtime": 60.6783,
|
| 2542 |
+
"eval_samples_per_second": 685.352,
|
| 2543 |
+
"eval_steps_per_second": 0.676,
|
| 2544 |
+
"step": 8450
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 3.9415719916531415,
|
| 2548 |
+
"grad_norm": 0.045575451105833054,
|
| 2549 |
+
"learning_rate": 5.558227567253832e-06,
|
| 2550 |
+
"loss": 0.1281,
|
| 2551 |
+
"step": 8500
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 3.9415719916531415,
|
| 2555 |
+
"eval_loss": 0.16032033338606583,
|
| 2556 |
+
"eval_runtime": 60.4563,
|
| 2557 |
+
"eval_samples_per_second": 687.868,
|
| 2558 |
+
"eval_steps_per_second": 0.678,
|
| 2559 |
+
"step": 8500
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 3.964757709251101,
|
| 2563 |
+
"grad_norm": 0.034972067922353745,
|
| 2564 |
+
"learning_rate": 5.200311285433213e-06,
|
| 2565 |
+
"loss": 0.1285,
|
| 2566 |
+
"step": 8550
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 3.964757709251101,
|
| 2570 |
+
"eval_loss": 0.1590997571686103,
|
| 2571 |
+
"eval_runtime": 60.7642,
|
| 2572 |
+
"eval_samples_per_second": 684.384,
|
| 2573 |
+
"eval_steps_per_second": 0.675,
|
| 2574 |
+
"step": 8550
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 3.987943426849061,
|
| 2578 |
+
"grad_norm": 0.05060684680938721,
|
| 2579 |
+
"learning_rate": 4.853673085668947e-06,
|
| 2580 |
+
"loss": 0.1293,
|
| 2581 |
+
"step": 8600
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 3.987943426849061,
|
| 2585 |
+
"eval_loss": 0.15924322809570868,
|
| 2586 |
+
"eval_runtime": 60.0799,
|
| 2587 |
+
"eval_samples_per_second": 692.178,
|
| 2588 |
+
"eval_steps_per_second": 0.682,
|
| 2589 |
+
"step": 8600
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 4.011129144447021,
|
| 2593 |
+
"grad_norm": 0.04898017644882202,
|
| 2594 |
+
"learning_rate": 4.5184002322740785e-06,
|
| 2595 |
+
"loss": 0.1283,
|
| 2596 |
+
"step": 8650
|
| 2597 |
+
},
|
| 2598 |
+
{
|
| 2599 |
+
"epoch": 4.011129144447021,
|
| 2600 |
+
"eval_loss": 0.1587491140112498,
|
| 2601 |
+
"eval_runtime": 60.6393,
|
| 2602 |
+
"eval_samples_per_second": 685.793,
|
| 2603 |
+
"eval_steps_per_second": 0.676,
|
| 2604 |
+
"step": 8650
|
| 2605 |
+
},
|
| 2606 |
+
{
|
| 2607 |
+
"epoch": 4.0343148620449805,
|
| 2608 |
+
"grad_norm": 0.058361586183309555,
|
| 2609 |
+
"learning_rate": 4.19457712839652e-06,
|
| 2610 |
+
"loss": 0.1277,
|
| 2611 |
+
"step": 8700
|
| 2612 |
+
},
|
| 2613 |
+
{
|
| 2614 |
+
"epoch": 4.0343148620449805,
|
| 2615 |
+
"eval_loss": 0.1597737118597627,
|
| 2616 |
+
"eval_runtime": 61.5486,
|
| 2617 |
+
"eval_samples_per_second": 675.661,
|
| 2618 |
+
"eval_steps_per_second": 0.666,
|
| 2619 |
+
"step": 8700
|
| 2620 |
+
},
|
| 2621 |
+
{
|
| 2622 |
+
"epoch": 4.05750057964294,
|
| 2623 |
+
"grad_norm": 0.05138258635997772,
|
| 2624 |
+
"learning_rate": 3.8822852947709375e-06,
|
| 2625 |
+
"loss": 0.1283,
|
| 2626 |
+
"step": 8750
|
| 2627 |
+
},
|
| 2628 |
+
{
|
| 2629 |
+
"epoch": 4.05750057964294,
|
| 2630 |
+
"eval_loss": 0.15985116115580386,
|
| 2631 |
+
"eval_runtime": 60.5634,
|
| 2632 |
+
"eval_samples_per_second": 686.652,
|
| 2633 |
+
"eval_steps_per_second": 0.677,
|
| 2634 |
+
"step": 8750
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 4.0806862972408995,
|
| 2638 |
+
"grad_norm": 0.0461881086230278,
|
| 2639 |
+
"learning_rate": 3.581603349196372e-06,
|
| 2640 |
+
"loss": 0.1288,
|
| 2641 |
+
"step": 8800
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 4.0806862972408995,
|
| 2645 |
+
"eval_loss": 0.15788726458515429,
|
| 2646 |
+
"eval_runtime": 60.6057,
|
| 2647 |
+
"eval_samples_per_second": 686.173,
|
| 2648 |
+
"eval_steps_per_second": 0.677,
|
| 2649 |
+
"step": 8800
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 4.103872014838859,
|
| 2653 |
+
"grad_norm": 0.0618111789226532,
|
| 2654 |
+
"learning_rate": 3.2926069867446675e-06,
|
| 2655 |
+
"loss": 0.1287,
|
| 2656 |
+
"step": 8850
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 4.103872014838859,
|
| 2660 |
+
"eval_loss": 0.15881183094974458,
|
| 2661 |
+
"eval_runtime": 60.3747,
|
| 2662 |
+
"eval_samples_per_second": 688.799,
|
| 2663 |
+
"eval_steps_per_second": 0.679,
|
| 2664 |
+
"step": 8850
|
| 2665 |
+
},
|
| 2666 |
+
{
|
| 2667 |
+
"epoch": 4.1270577324368185,
|
| 2668 |
+
"grad_norm": 0.04804789274930954,
|
| 2669 |
+
"learning_rate": 3.0153689607045845e-06,
|
| 2670 |
+
"loss": 0.1294,
|
| 2671 |
+
"step": 8900
|
| 2672 |
+
},
|
| 2673 |
+
{
|
| 2674 |
+
"epoch": 4.1270577324368185,
|
| 2675 |
+
"eval_loss": 0.1607356553004913,
|
| 2676 |
+
"eval_runtime": 60.6979,
|
| 2677 |
+
"eval_samples_per_second": 685.131,
|
| 2678 |
+
"eval_steps_per_second": 0.675,
|
| 2679 |
+
"step": 8900
|
| 2680 |
+
},
|
| 2681 |
+
{
|
| 2682 |
+
"epoch": 4.150243450034779,
|
| 2683 |
+
"grad_norm": 0.04835003986954689,
|
| 2684 |
+
"learning_rate": 2.7499590642665774e-06,
|
| 2685 |
+
"loss": 0.1277,
|
| 2686 |
+
"step": 8950
|
| 2687 |
+
},
|
| 2688 |
+
{
|
| 2689 |
+
"epoch": 4.150243450034779,
|
| 2690 |
+
"eval_loss": 0.1598761189516689,
|
| 2691 |
+
"eval_runtime": 61.2003,
|
| 2692 |
+
"eval_samples_per_second": 679.507,
|
| 2693 |
+
"eval_steps_per_second": 0.67,
|
| 2694 |
+
"step": 8950
|
| 2695 |
+
},
|
| 2696 |
+
{
|
| 2697 |
+
"epoch": 4.173429167632738,
|
| 2698 |
+
"grad_norm": 0.05750919133424759,
|
| 2699 |
+
"learning_rate": 2.496444112952734e-06,
|
| 2700 |
+
"loss": 0.1285,
|
| 2701 |
+
"step": 9000
|
| 2702 |
+
},
|
| 2703 |
+
{
|
| 2704 |
+
"epoch": 4.173429167632738,
|
| 2705 |
+
"eval_loss": 0.15946166188972705,
|
| 2706 |
+
"eval_runtime": 60.6795,
|
| 2707 |
+
"eval_samples_per_second": 685.339,
|
| 2708 |
+
"eval_steps_per_second": 0.676,
|
| 2709 |
+
"step": 9000
|
| 2710 |
+
},
|
| 2711 |
+
{
|
| 2712 |
+
"epoch": 4.196614885230698,
|
| 2713 |
+
"grad_norm": 0.06801807135343552,
|
| 2714 |
+
"learning_rate": 2.2548879277963064e-06,
|
| 2715 |
+
"loss": 0.1289,
|
| 2716 |
+
"step": 9050
|
| 2717 |
+
},
|
| 2718 |
+
{
|
| 2719 |
+
"epoch": 4.196614885230698,
|
| 2720 |
+
"eval_loss": 0.1609577221237089,
|
| 2721 |
+
"eval_runtime": 61.0186,
|
| 2722 |
+
"eval_samples_per_second": 681.53,
|
| 2723 |
+
"eval_steps_per_second": 0.672,
|
| 2724 |
+
"step": 9050
|
| 2725 |
+
},
|
| 2726 |
+
{
|
| 2727 |
+
"epoch": 4.219800602828657,
|
| 2728 |
+
"grad_norm": 0.04383298382163048,
|
| 2729 |
+
"learning_rate": 2.0253513192751373e-06,
|
| 2730 |
+
"loss": 0.1289,
|
| 2731 |
+
"step": 9100
|
| 2732 |
+
},
|
| 2733 |
+
{
|
| 2734 |
+
"epoch": 4.219800602828657,
|
| 2735 |
+
"eval_loss": 0.1598739506376352,
|
| 2736 |
+
"eval_runtime": 60.6256,
|
| 2737 |
+
"eval_samples_per_second": 685.948,
|
| 2738 |
+
"eval_steps_per_second": 0.676,
|
| 2739 |
+
"step": 9100
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 4.242986320426617,
|
| 2743 |
+
"grad_norm": 0.044339120388031006,
|
| 2744 |
+
"learning_rate": 1.807892072002898e-06,
|
| 2745 |
+
"loss": 0.1283,
|
| 2746 |
+
"step": 9150
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 4.242986320426617,
|
| 2750 |
+
"eval_loss": 0.158920794598519,
|
| 2751 |
+
"eval_runtime": 60.5454,
|
| 2752 |
+
"eval_samples_per_second": 686.856,
|
| 2753 |
+
"eval_steps_per_second": 0.677,
|
| 2754 |
+
"step": 9150
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 4.2661720380245765,
|
| 2758 |
+
"grad_norm": 0.04090524837374687,
|
| 2759 |
+
"learning_rate": 1.6025649301821876e-06,
|
| 2760 |
+
"loss": 0.1282,
|
| 2761 |
+
"step": 9200
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 4.2661720380245765,
|
| 2765 |
+
"eval_loss": 0.1596859048948022,
|
| 2766 |
+
"eval_runtime": 60.7716,
|
| 2767 |
+
"eval_samples_per_second": 684.3,
|
| 2768 |
+
"eval_steps_per_second": 0.675,
|
| 2769 |
+
"step": 9200
|
| 2770 |
+
},
|
| 2771 |
+
{
|
| 2772 |
+
"epoch": 4.289357755622537,
|
| 2773 |
+
"grad_norm": 0.042642634361982346,
|
| 2774 |
+
"learning_rate": 1.4094215838229176e-06,
|
| 2775 |
+
"loss": 0.1286,
|
| 2776 |
+
"step": 9250
|
| 2777 |
+
},
|
| 2778 |
+
{
|
| 2779 |
+
"epoch": 4.289357755622537,
|
| 2780 |
+
"eval_loss": 0.16079005979316527,
|
| 2781 |
+
"eval_runtime": 60.6239,
|
| 2782 |
+
"eval_samples_per_second": 685.967,
|
| 2783 |
+
"eval_steps_per_second": 0.676,
|
| 2784 |
+
"step": 9250
|
| 2785 |
+
},
|
| 2786 |
+
{
|
| 2787 |
+
"epoch": 4.312543473220496,
|
| 2788 |
+
"grad_norm": 0.04924129322171211,
|
| 2789 |
+
"learning_rate": 1.2285106557296477e-06,
|
| 2790 |
+
"loss": 0.1287,
|
| 2791 |
+
"step": 9300
|
| 2792 |
+
},
|
| 2793 |
+
{
|
| 2794 |
+
"epoch": 4.312543473220496,
|
| 2795 |
+
"eval_loss": 0.16084020796323667,
|
| 2796 |
+
"eval_runtime": 60.2581,
|
| 2797 |
+
"eval_samples_per_second": 690.131,
|
| 2798 |
+
"eval_steps_per_second": 0.68,
|
| 2799 |
+
"step": 9300
|
| 2800 |
+
},
|
| 2801 |
+
{
|
| 2802 |
+
"epoch": 4.335729190818456,
|
| 2803 |
+
"grad_norm": 0.04222133755683899,
|
| 2804 |
+
"learning_rate": 1.0598776892610685e-06,
|
| 2805 |
+
"loss": 0.1287,
|
| 2806 |
+
"step": 9350
|
| 2807 |
+
},
|
| 2808 |
+
{
|
| 2809 |
+
"epoch": 4.335729190818456,
|
| 2810 |
+
"eval_loss": 0.16020395921618655,
|
| 2811 |
+
"eval_runtime": 60.5816,
|
| 2812 |
+
"eval_samples_per_second": 686.446,
|
| 2813 |
+
"eval_steps_per_second": 0.677,
|
| 2814 |
+
"step": 9350
|
| 2815 |
+
},
|
| 2816 |
+
{
|
| 2817 |
+
"epoch": 4.358914908416415,
|
| 2818 |
+
"grad_norm": 0.05593874678015709,
|
| 2819 |
+
"learning_rate": 9.035651368646648e-07,
|
| 2820 |
+
"loss": 0.1286,
|
| 2821 |
+
"step": 9400
|
| 2822 |
+
},
|
| 2823 |
+
{
|
| 2824 |
+
"epoch": 4.358914908416415,
|
| 2825 |
+
"eval_loss": 0.15957607987630548,
|
| 2826 |
+
"eval_runtime": 60.8726,
|
| 2827 |
+
"eval_samples_per_second": 683.164,
|
| 2828 |
+
"eval_steps_per_second": 0.674,
|
| 2829 |
+
"step": 9400
|
| 2830 |
+
},
|
| 2831 |
+
{
|
| 2832 |
+
"epoch": 4.382100626014375,
|
| 2833 |
+
"grad_norm": 0.059049129486083984,
|
| 2834 |
+
"learning_rate": 7.596123493895991e-07,
|
| 2835 |
+
"loss": 0.1289,
|
| 2836 |
+
"step": 9450
|
| 2837 |
+
},
|
| 2838 |
+
{
|
| 2839 |
+
"epoch": 4.382100626014375,
|
| 2840 |
+
"eval_loss": 0.15975197211451994,
|
| 2841 |
+
"eval_runtime": 60.6704,
|
| 2842 |
+
"eval_samples_per_second": 685.441,
|
| 2843 |
+
"eval_steps_per_second": 0.676,
|
| 2844 |
+
"step": 9450
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 4.405286343612334,
|
| 2848 |
+
"grad_norm": 0.053555767983198166,
|
| 2849 |
+
"learning_rate": 6.280555661802856e-07,
|
| 2850 |
+
"loss": 0.1286,
|
| 2851 |
+
"step": 9500
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 4.405286343612334,
|
| 2855 |
+
"eval_loss": 0.16117730336557945,
|
| 2856 |
+
"eval_runtime": 61.9214,
|
| 2857 |
+
"eval_samples_per_second": 671.593,
|
| 2858 |
+
"eval_steps_per_second": 0.662,
|
| 2859 |
+
"step": 9500
|
| 2860 |
+
},
|
| 2861 |
+
{
|
| 2862 |
+
"epoch": 4.428472061210295,
|
| 2863 |
+
"grad_norm": 0.04488294571638107,
|
| 2864 |
+
"learning_rate": 5.089279059533658e-07,
|
| 2865 |
+
"loss": 0.1281,
|
| 2866 |
+
"step": 9550
|
| 2867 |
+
},
|
| 2868 |
+
{
|
| 2869 |
+
"epoch": 4.428472061210295,
|
| 2870 |
+
"eval_loss": 0.15896389365558133,
|
| 2871 |
+
"eval_runtime": 62.0889,
|
| 2872 |
+
"eval_samples_per_second": 669.782,
|
| 2873 |
+
"eval_steps_per_second": 0.66,
|
| 2874 |
+
"step": 9550
|
| 2875 |
+
},
|
| 2876 |
+
{
|
| 2877 |
+
"epoch": 4.451657778808254,
|
| 2878 |
+
"grad_norm": 0.044143371284008026,
|
| 2879 |
+
"learning_rate": 4.02259358460233e-07,
|
| 2880 |
+
"loss": 0.1276,
|
| 2881 |
+
"step": 9600
|
| 2882 |
+
},
|
| 2883 |
+
{
|
| 2884 |
+
"epoch": 4.451657778808254,
|
| 2885 |
+
"eval_loss": 0.15880485748262804,
|
| 2886 |
+
"eval_runtime": 61.9007,
|
| 2887 |
+
"eval_samples_per_second": 671.818,
|
| 2888 |
+
"eval_steps_per_second": 0.662,
|
| 2889 |
+
"step": 9600
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 4.474843496406214,
|
| 2893 |
+
"grad_norm": 0.054890409111976624,
|
| 2894 |
+
"learning_rate": 3.080767769372939e-07,
|
| 2895 |
+
"loss": 0.1289,
|
| 2896 |
+
"step": 9650
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 4.474843496406214,
|
| 2900 |
+
"eval_loss": 0.15899979579394047,
|
| 2901 |
+
"eval_runtime": 61.7264,
|
| 2902 |
+
"eval_samples_per_second": 673.714,
|
| 2903 |
+
"eval_steps_per_second": 0.664,
|
| 2904 |
+
"step": 9650
|
| 2905 |
+
},
|
| 2906 |
+
{
|
| 2907 |
+
"epoch": 4.498029214004173,
|
| 2908 |
+
"grad_norm": 0.04276006668806076,
|
| 2909 |
+
"learning_rate": 2.2640387134577058e-07,
|
| 2910 |
+
"loss": 0.1284,
|
| 2911 |
+
"step": 9700
|
| 2912 |
+
},
|
| 2913 |
+
{
|
| 2914 |
+
"epoch": 4.498029214004173,
|
| 2915 |
+
"eval_loss": 0.1587265635928511,
|
| 2916 |
+
"eval_runtime": 61.254,
|
| 2917 |
+
"eval_samples_per_second": 678.911,
|
| 2918 |
+
"eval_steps_per_second": 0.669,
|
| 2919 |
+
"step": 9700
|
| 2920 |
+
},
|
| 2921 |
+
{
|
| 2922 |
+
"epoch": 4.521214931602133,
|
| 2923 |
+
"grad_norm": 0.04374442994594574,
|
| 2924 |
+
"learning_rate": 1.5726120240288634e-07,
|
| 2925 |
+
"loss": 0.1284,
|
| 2926 |
+
"step": 9750
|
| 2927 |
+
},
|
| 2928 |
+
{
|
| 2929 |
+
"epoch": 4.521214931602133,
|
| 2930 |
+
"eval_loss": 0.1596641951113874,
|
| 2931 |
+
"eval_runtime": 61.5999,
|
| 2932 |
+
"eval_samples_per_second": 675.099,
|
| 2933 |
+
"eval_steps_per_second": 0.666,
|
| 2934 |
+
"step": 9750
|
| 2935 |
+
},
|
| 2936 |
+
{
|
| 2937 |
+
"epoch": 4.544400649200092,
|
| 2938 |
+
"grad_norm": 0.039518803358078,
|
| 2939 |
+
"learning_rate": 1.0066617640578368e-07,
|
| 2940 |
+
"loss": 0.1297,
|
| 2941 |
+
"step": 9800
|
| 2942 |
+
},
|
| 2943 |
+
{
|
| 2944 |
+
"epoch": 4.544400649200092,
|
| 2945 |
+
"eval_loss": 0.15941302591091938,
|
| 2946 |
+
"eval_runtime": 61.7763,
|
| 2947 |
+
"eval_samples_per_second": 673.17,
|
| 2948 |
+
"eval_steps_per_second": 0.664,
|
| 2949 |
+
"step": 9800
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 4.567586366798053,
|
| 2953 |
+
"grad_norm": 0.037454187870025635,
|
| 2954 |
+
"learning_rate": 5.663304084960186e-08,
|
| 2955 |
+
"loss": 0.1276,
|
| 2956 |
+
"step": 9850
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 4.567586366798053,
|
| 2960 |
+
"eval_loss": 0.15932704533345807,
|
| 2961 |
+
"eval_runtime": 61.0673,
|
| 2962 |
+
"eval_samples_per_second": 680.987,
|
| 2963 |
+
"eval_steps_per_second": 0.671,
|
| 2964 |
+
"step": 9850
|
| 2965 |
+
},
|
| 2966 |
+
{
|
| 2967 |
+
"epoch": 4.590772084396012,
|
| 2968 |
+
"grad_norm": 0.05642937496304512,
|
| 2969 |
+
"learning_rate": 2.5172880840745873e-08,
|
| 2970 |
+
"loss": 0.129,
|
| 2971 |
+
"step": 9900
|
| 2972 |
+
},
|
| 2973 |
+
{
|
| 2974 |
+
"epoch": 4.590772084396012,
|
| 2975 |
+
"eval_loss": 0.15923822751292327,
|
| 2976 |
+
"eval_runtime": 61.7949,
|
| 2977 |
+
"eval_samples_per_second": 672.968,
|
| 2978 |
+
"eval_steps_per_second": 0.663,
|
| 2979 |
+
"step": 9900
|
| 2980 |
+
},
|
| 2981 |
+
{
|
| 2982 |
+
"epoch": 4.613957801993972,
|
| 2983 |
+
"grad_norm": 0.03662274032831192,
|
| 2984 |
+
"learning_rate": 6.293616306246586e-09,
|
| 2985 |
+
"loss": 0.1285,
|
| 2986 |
+
"step": 9950
|
| 2987 |
+
},
|
| 2988 |
+
{
|
| 2989 |
+
"epoch": 4.613957801993972,
|
| 2990 |
+
"eval_loss": 0.160277338388973,
|
| 2991 |
+
"eval_runtime": 62.2438,
|
| 2992 |
+
"eval_samples_per_second": 668.115,
|
| 2993 |
+
"eval_steps_per_second": 0.659,
|
| 2994 |
+
"step": 9950
|
| 2995 |
+
},
|
| 2996 |
+
{
|
| 2997 |
+
"epoch": 4.637143519591931,
|
| 2998 |
+
"grad_norm": 0.0563049279153347,
|
| 2999 |
+
"learning_rate": 0.0,
|
| 3000 |
+
"loss": 0.1282,
|
| 3001 |
+
"step": 10000
|
| 3002 |
+
},
|
| 3003 |
+
{
|
| 3004 |
+
"epoch": 4.637143519591931,
|
| 3005 |
+
"eval_loss": 0.16006394581914293,
|
| 3006 |
+
"eval_runtime": 61.4917,
|
| 3007 |
+
"eval_samples_per_second": 676.286,
|
| 3008 |
+
"eval_steps_per_second": 0.667,
|
| 3009 |
+
"step": 10000
|
| 3010 |
+
},
|
| 3011 |
+
{
|
| 3012 |
+
"epoch": 4.637143519591931,
|
| 3013 |
+
"step": 10000,
|
| 3014 |
+
"total_flos": 2.3231400526217216e+17,
|
| 3015 |
+
"train_loss": 0.13326009378433226,
|
| 3016 |
+
"train_runtime": 41368.3669,
|
| 3017 |
+
"train_samples_per_second": 495.064,
|
| 3018 |
+
"train_steps_per_second": 0.242
|
| 3019 |
+
}
|
| 3020 |
+
],
|
| 3021 |
+
"logging_steps": 50,
|
| 3022 |
+
"max_steps": 10000,
|
| 3023 |
+
"num_input_tokens_seen": 0,
|
| 3024 |
+
"num_train_epochs": 5,
|
| 3025 |
+
"save_steps": 50,
|
| 3026 |
+
"total_flos": 2.3231400526217216e+17,
|
| 3027 |
+
"train_batch_size": 1024,
|
| 3028 |
+
"trial_name": null,
|
| 3029 |
+
"trial_params": null
|
| 3030 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3df77597c69632ab7d5a5d7987fd817d663727a57d77ca069bc92a72238da772
|
| 3 |
+
size 5393
|