Commit
•
8c8938e
1
Parent(s):
c11702d
Evaluation on the test set completed on 2024_06_03.
Browse files- README.md +175 -0
- all_results.json +17 -0
- logs/events.out.tfevents.1717167376.datavisu2 +2 -2
- logs/events.out.tfevents.1717405982.datavisu2 +3 -0
- model.safetensors +1 -1
- test_results.json +12 -0
- train_results.json +9 -0
- trainer_state.json +1915 -0
README.md
ADDED
|
@@ -0,0 +1,175 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: facebook/dinov2-large
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
model-index:
|
| 9 |
+
- name: DinoVdeau-large-2024_05_31-batch-size32_epochs150_freeze
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# DinoVdeau-large-2024_05_31-batch-size32_epochs150_freeze
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [facebook/dinov2-large](https://huggingface.co/facebook/dinov2-large) on the None dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.1235
|
| 21 |
+
- F1 Micro: 0.8217
|
| 22 |
+
- F1 Macro: 0.7173
|
| 23 |
+
- Roc Auc: 0.8829
|
| 24 |
+
- Accuracy: 0.3183
|
| 25 |
+
- Learning Rate: 0.0000
|
| 26 |
+
|
| 27 |
+
## Model description
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Intended uses & limitations
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training and evaluation data
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Training procedure
|
| 40 |
+
|
| 41 |
+
### Training hyperparameters
|
| 42 |
+
|
| 43 |
+
The following hyperparameters were used during training:
|
| 44 |
+
- learning_rate: 0.001
|
| 45 |
+
- train_batch_size: 32
|
| 46 |
+
- eval_batch_size: 32
|
| 47 |
+
- seed: 42
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: linear
|
| 50 |
+
- num_epochs: 150
|
| 51 |
+
- mixed_precision_training: Native AMP
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | Roc Auc | Accuracy | Rate |
|
| 56 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:|:-------:|:--------:|:------:|
|
| 57 |
+
| No log | 1.0 | 273 | 0.1727 | 0.7472 | 0.5105 | 0.8326 | 0.2372 | 0.001 |
|
| 58 |
+
| 0.2751 | 2.0 | 546 | 0.1584 | 0.7591 | 0.5755 | 0.8364 | 0.2506 | 0.001 |
|
| 59 |
+
| 0.2751 | 3.0 | 819 | 0.1513 | 0.7712 | 0.5998 | 0.8511 | 0.2468 | 0.001 |
|
| 60 |
+
| 0.1714 | 4.0 | 1092 | 0.1497 | 0.7649 | 0.6103 | 0.8344 | 0.2516 | 0.001 |
|
| 61 |
+
| 0.1714 | 5.0 | 1365 | 0.1479 | 0.7743 | 0.6048 | 0.8514 | 0.2619 | 0.001 |
|
| 62 |
+
| 0.1664 | 6.0 | 1638 | 0.1465 | 0.7744 | 0.6142 | 0.8476 | 0.2386 | 0.001 |
|
| 63 |
+
| 0.1664 | 7.0 | 1911 | 0.1471 | 0.7746 | 0.6379 | 0.8434 | 0.2523 | 0.001 |
|
| 64 |
+
| 0.1613 | 8.0 | 2184 | 0.1488 | 0.7730 | 0.6350 | 0.8432 | 0.2492 | 0.001 |
|
| 65 |
+
| 0.1613 | 9.0 | 2457 | 0.1696 | 0.7571 | 0.5914 | 0.8382 | 0.1708 | 0.001 |
|
| 66 |
+
| 0.1607 | 10.0 | 2730 | 0.1497 | 0.7753 | 0.6155 | 0.8459 | 0.2633 | 0.001 |
|
| 67 |
+
| 0.1589 | 11.0 | 3003 | 0.1469 | 0.7747 | 0.6157 | 0.8483 | 0.2592 | 0.001 |
|
| 68 |
+
| 0.1589 | 12.0 | 3276 | 0.1456 | 0.7830 | 0.6489 | 0.8598 | 0.2558 | 0.001 |
|
| 69 |
+
| 0.1578 | 13.0 | 3549 | 0.1428 | 0.7843 | 0.6581 | 0.8557 | 0.2554 | 0.001 |
|
| 70 |
+
| 0.1578 | 14.0 | 3822 | 0.1435 | 0.7832 | 0.6419 | 0.8542 | 0.2506 | 0.001 |
|
| 71 |
+
| 0.158 | 15.0 | 4095 | 0.1454 | 0.7811 | 0.6573 | 0.8563 | 0.2534 | 0.001 |
|
| 72 |
+
| 0.158 | 16.0 | 4368 | 0.1448 | 0.7866 | 0.6504 | 0.8603 | 0.2640 | 0.001 |
|
| 73 |
+
| 0.1587 | 17.0 | 4641 | 0.1431 | 0.7819 | 0.6441 | 0.8543 | 0.2595 | 0.001 |
|
| 74 |
+
| 0.1587 | 18.0 | 4914 | 0.1447 | 0.7839 | 0.6413 | 0.8548 | 0.2547 | 0.001 |
|
| 75 |
+
| 0.1571 | 19.0 | 5187 | 0.1415 | 0.7836 | 0.6442 | 0.8506 | 0.2647 | 0.001 |
|
| 76 |
+
| 0.1571 | 20.0 | 5460 | 0.1406 | 0.7839 | 0.6437 | 0.8513 | 0.2609 | 0.001 |
|
| 77 |
+
| 0.1577 | 21.0 | 5733 | 0.1414 | 0.7847 | 0.6461 | 0.8527 | 0.2575 | 0.001 |
|
| 78 |
+
| 0.1556 | 22.0 | 6006 | 0.1386 | 0.7884 | 0.6587 | 0.8567 | 0.2619 | 0.001 |
|
| 79 |
+
| 0.1556 | 23.0 | 6279 | 0.1382 | 0.7900 | 0.6503 | 0.8567 | 0.2681 | 0.001 |
|
| 80 |
+
| 0.1558 | 24.0 | 6552 | 0.1389 | 0.7881 | 0.6519 | 0.8561 | 0.2650 | 0.001 |
|
| 81 |
+
| 0.1558 | 25.0 | 6825 | 0.1397 | 0.7883 | 0.6554 | 0.8556 | 0.2695 | 0.001 |
|
| 82 |
+
| 0.1551 | 26.0 | 7098 | 0.1401 | 0.7794 | 0.6350 | 0.8453 | 0.2650 | 0.001 |
|
| 83 |
+
| 0.1551 | 27.0 | 7371 | 0.1402 | 0.7915 | 0.6668 | 0.8638 | 0.2499 | 0.001 |
|
| 84 |
+
| 0.156 | 28.0 | 7644 | 0.1389 | 0.7908 | 0.6576 | 0.8581 | 0.2640 | 0.001 |
|
| 85 |
+
| 0.156 | 29.0 | 7917 | 0.1401 | 0.7940 | 0.6657 | 0.8654 | 0.2681 | 0.001 |
|
| 86 |
+
| 0.155 | 30.0 | 8190 | 0.1346 | 0.7987 | 0.6711 | 0.8669 | 0.2740 | 0.0001 |
|
| 87 |
+
| 0.155 | 31.0 | 8463 | 0.1324 | 0.7978 | 0.6774 | 0.8613 | 0.2784 | 0.0001 |
|
| 88 |
+
| 0.1466 | 32.0 | 8736 | 0.1333 | 0.8019 | 0.6779 | 0.8704 | 0.2729 | 0.0001 |
|
| 89 |
+
| 0.1431 | 33.0 | 9009 | 0.1317 | 0.8034 | 0.6849 | 0.8673 | 0.2795 | 0.0001 |
|
| 90 |
+
| 0.1431 | 34.0 | 9282 | 0.1307 | 0.8021 | 0.6836 | 0.8660 | 0.2798 | 0.0001 |
|
| 91 |
+
| 0.1413 | 35.0 | 9555 | 0.1303 | 0.8046 | 0.6903 | 0.8719 | 0.2853 | 0.0001 |
|
| 92 |
+
| 0.1413 | 36.0 | 9828 | 0.1297 | 0.8051 | 0.6932 | 0.8664 | 0.2901 | 0.0001 |
|
| 93 |
+
| 0.14 | 37.0 | 10101 | 0.1296 | 0.8035 | 0.6885 | 0.8671 | 0.2884 | 0.0001 |
|
| 94 |
+
| 0.14 | 38.0 | 10374 | 0.1293 | 0.8043 | 0.6888 | 0.8673 | 0.2870 | 0.0001 |
|
| 95 |
+
| 0.1391 | 39.0 | 10647 | 0.1304 | 0.8033 | 0.6870 | 0.8645 | 0.2877 | 0.0001 |
|
| 96 |
+
| 0.1391 | 40.0 | 10920 | 0.1289 | 0.8064 | 0.6935 | 0.8713 | 0.2908 | 0.0001 |
|
| 97 |
+
| 0.1375 | 41.0 | 11193 | 0.1298 | 0.8037 | 0.6924 | 0.8650 | 0.2870 | 0.0001 |
|
| 98 |
+
| 0.1375 | 42.0 | 11466 | 0.1285 | 0.8086 | 0.6900 | 0.8722 | 0.2960 | 0.0001 |
|
| 99 |
+
| 0.1369 | 43.0 | 11739 | 0.1285 | 0.8067 | 0.6898 | 0.8708 | 0.2870 | 0.0001 |
|
| 100 |
+
| 0.1356 | 44.0 | 12012 | 0.1286 | 0.8078 | 0.6931 | 0.8706 | 0.2922 | 0.0001 |
|
| 101 |
+
| 0.1356 | 45.0 | 12285 | 0.1276 | 0.8107 | 0.7037 | 0.8752 | 0.2956 | 0.0001 |
|
| 102 |
+
| 0.1348 | 46.0 | 12558 | 0.1278 | 0.8078 | 0.6951 | 0.8703 | 0.2956 | 0.0001 |
|
| 103 |
+
| 0.1348 | 47.0 | 12831 | 0.1277 | 0.8114 | 0.7063 | 0.8774 | 0.2919 | 0.0001 |
|
| 104 |
+
| 0.1343 | 48.0 | 13104 | 0.1276 | 0.8052 | 0.6955 | 0.8649 | 0.2908 | 0.0001 |
|
| 105 |
+
| 0.1343 | 49.0 | 13377 | 0.1278 | 0.8099 | 0.7021 | 0.8740 | 0.2884 | 0.0001 |
|
| 106 |
+
| 0.1344 | 50.0 | 13650 | 0.1270 | 0.8102 | 0.7060 | 0.8725 | 0.2939 | 0.0001 |
|
| 107 |
+
| 0.1344 | 51.0 | 13923 | 0.1274 | 0.8129 | 0.7098 | 0.8780 | 0.2977 | 0.0001 |
|
| 108 |
+
| 0.1331 | 52.0 | 14196 | 0.1274 | 0.8122 | 0.7076 | 0.8765 | 0.2919 | 0.0001 |
|
| 109 |
+
| 0.1331 | 53.0 | 14469 | 0.1278 | 0.8084 | 0.6942 | 0.8700 | 0.2932 | 0.0001 |
|
| 110 |
+
| 0.1325 | 54.0 | 14742 | 0.1264 | 0.8096 | 0.6945 | 0.8717 | 0.2939 | 0.0001 |
|
| 111 |
+
| 0.1318 | 55.0 | 15015 | 0.1261 | 0.8117 | 0.7114 | 0.8724 | 0.2984 | 0.0001 |
|
| 112 |
+
| 0.1318 | 56.0 | 15288 | 0.1269 | 0.8113 | 0.7063 | 0.8734 | 0.2991 | 0.0001 |
|
| 113 |
+
| 0.1311 | 57.0 | 15561 | 0.1260 | 0.8135 | 0.7093 | 0.8757 | 0.2991 | 0.0001 |
|
| 114 |
+
| 0.1311 | 58.0 | 15834 | 0.1269 | 0.8129 | 0.7038 | 0.8764 | 0.2925 | 0.0001 |
|
| 115 |
+
| 0.1307 | 59.0 | 16107 | 0.1266 | 0.8112 | 0.7074 | 0.8727 | 0.2922 | 0.0001 |
|
| 116 |
+
| 0.1307 | 60.0 | 16380 | 0.1264 | 0.8091 | 0.6985 | 0.8704 | 0.2932 | 0.0001 |
|
| 117 |
+
| 0.1303 | 61.0 | 16653 | 0.1258 | 0.8097 | 0.7041 | 0.8680 | 0.2960 | 0.0001 |
|
| 118 |
+
| 0.1303 | 62.0 | 16926 | 0.1270 | 0.8133 | 0.7066 | 0.8771 | 0.2912 | 0.0001 |
|
| 119 |
+
| 0.1299 | 63.0 | 17199 | 0.1257 | 0.8111 | 0.7060 | 0.8692 | 0.2970 | 0.0001 |
|
| 120 |
+
| 0.1299 | 64.0 | 17472 | 0.1253 | 0.8139 | 0.7090 | 0.8757 | 0.2939 | 0.0001 |
|
| 121 |
+
| 0.1284 | 65.0 | 17745 | 0.1251 | 0.8127 | 0.7047 | 0.8726 | 0.2963 | 0.0001 |
|
| 122 |
+
| 0.1283 | 66.0 | 18018 | 0.1256 | 0.8140 | 0.7119 | 0.8767 | 0.2949 | 0.0001 |
|
| 123 |
+
| 0.1283 | 67.0 | 18291 | 0.1254 | 0.8111 | 0.7062 | 0.8702 | 0.2939 | 0.0001 |
|
| 124 |
+
| 0.1281 | 68.0 | 18564 | 0.1256 | 0.8152 | 0.7113 | 0.8782 | 0.2960 | 0.0001 |
|
| 125 |
+
| 0.1281 | 69.0 | 18837 | 0.1258 | 0.8143 | 0.7141 | 0.8767 | 0.2994 | 0.0001 |
|
| 126 |
+
| 0.1274 | 70.0 | 19110 | 0.1248 | 0.8162 | 0.7184 | 0.8781 | 0.2984 | 0.0001 |
|
| 127 |
+
| 0.1274 | 71.0 | 19383 | 0.1255 | 0.8134 | 0.7097 | 0.8746 | 0.2963 | 0.0001 |
|
| 128 |
+
| 0.1274 | 72.0 | 19656 | 0.1268 | 0.8133 | 0.7067 | 0.8783 | 0.2980 | 0.0001 |
|
| 129 |
+
| 0.1274 | 73.0 | 19929 | 0.1294 | 0.8154 | 0.7201 | 0.8787 | 0.3004 | 0.0001 |
|
| 130 |
+
| 0.1272 | 74.0 | 20202 | 0.1252 | 0.8141 | 0.7157 | 0.8753 | 0.2960 | 0.0001 |
|
| 131 |
+
| 0.1272 | 75.0 | 20475 | 0.1253 | 0.8165 | 0.7162 | 0.8776 | 0.2998 | 0.0001 |
|
| 132 |
+
| 0.1265 | 76.0 | 20748 | 0.1251 | 0.8126 | 0.7051 | 0.8738 | 0.2980 | 0.0001 |
|
| 133 |
+
| 0.1253 | 77.0 | 21021 | 0.1238 | 0.8170 | 0.7199 | 0.8783 | 0.3018 | 1e-05 |
|
| 134 |
+
| 0.1253 | 78.0 | 21294 | 0.1243 | 0.8142 | 0.7125 | 0.8738 | 0.2998 | 1e-05 |
|
| 135 |
+
| 0.1231 | 79.0 | 21567 | 0.1235 | 0.8182 | 0.7211 | 0.8800 | 0.3029 | 1e-05 |
|
| 136 |
+
| 0.1231 | 80.0 | 21840 | 0.1240 | 0.8199 | 0.7272 | 0.8845 | 0.3022 | 1e-05 |
|
| 137 |
+
| 0.1232 | 81.0 | 22113 | 0.1242 | 0.8163 | 0.7214 | 0.8763 | 0.3011 | 1e-05 |
|
| 138 |
+
| 0.1232 | 82.0 | 22386 | 0.1238 | 0.8184 | 0.7221 | 0.8806 | 0.3042 | 1e-05 |
|
| 139 |
+
| 0.1223 | 83.0 | 22659 | 0.1235 | 0.8180 | 0.7213 | 0.8794 | 0.3018 | 1e-05 |
|
| 140 |
+
| 0.1223 | 84.0 | 22932 | 0.1237 | 0.8177 | 0.7233 | 0.8814 | 0.3025 | 1e-05 |
|
| 141 |
+
| 0.1216 | 85.0 | 23205 | 0.1232 | 0.8164 | 0.7142 | 0.8759 | 0.3046 | 1e-05 |
|
| 142 |
+
| 0.1216 | 86.0 | 23478 | 0.1237 | 0.8165 | 0.7243 | 0.8763 | 0.3035 | 1e-05 |
|
| 143 |
+
| 0.1217 | 87.0 | 23751 | 0.1235 | 0.8181 | 0.7201 | 0.8787 | 0.3022 | 1e-05 |
|
| 144 |
+
| 0.1206 | 88.0 | 24024 | 0.1239 | 0.8188 | 0.7209 | 0.8824 | 0.3035 | 1e-05 |
|
| 145 |
+
| 0.1206 | 89.0 | 24297 | 0.1237 | 0.8176 | 0.7172 | 0.8786 | 0.3029 | 1e-05 |
|
| 146 |
+
| 0.1215 | 90.0 | 24570 | 0.1232 | 0.8170 | 0.7203 | 0.8768 | 0.3059 | 1e-05 |
|
| 147 |
+
| 0.1215 | 91.0 | 24843 | 0.1235 | 0.8199 | 0.7244 | 0.8821 | 0.3035 | 1e-05 |
|
| 148 |
+
| 0.1216 | 92.0 | 25116 | 0.1241 | 0.8201 | 0.7264 | 0.8828 | 0.3056 | 1e-05 |
|
| 149 |
+
| 0.1216 | 93.0 | 25389 | 0.1233 | 0.8192 | 0.7266 | 0.8824 | 0.3029 | 1e-05 |
|
| 150 |
+
| 0.1203 | 94.0 | 25662 | 0.1240 | 0.8148 | 0.7139 | 0.8757 | 0.2977 | 1e-05 |
|
| 151 |
+
| 0.1203 | 95.0 | 25935 | 0.1237 | 0.8203 | 0.7273 | 0.8847 | 0.2980 | 1e-05 |
|
| 152 |
+
| 0.1215 | 96.0 | 26208 | 0.1237 | 0.8193 | 0.7203 | 0.8807 | 0.3049 | 1e-05 |
|
| 153 |
+
| 0.1215 | 97.0 | 26481 | 0.1235 | 0.8186 | 0.7213 | 0.8818 | 0.3008 | 0.0000 |
|
| 154 |
+
| 0.1211 | 98.0 | 26754 | 0.1231 | 0.8168 | 0.7173 | 0.8775 | 0.3039 | 0.0000 |
|
| 155 |
+
| 0.1202 | 99.0 | 27027 | 0.1234 | 0.8195 | 0.7266 | 0.8828 | 0.3022 | 0.0000 |
|
| 156 |
+
| 0.1202 | 100.0 | 27300 | 0.1234 | 0.8177 | 0.7212 | 0.8792 | 0.3025 | 0.0000 |
|
| 157 |
+
| 0.1202 | 101.0 | 27573 | 0.1230 | 0.8196 | 0.7275 | 0.8822 | 0.3008 | 0.0000 |
|
| 158 |
+
| 0.1202 | 102.0 | 27846 | 0.1234 | 0.8154 | 0.7137 | 0.8739 | 0.3015 | 0.0000 |
|
| 159 |
+
| 0.1212 | 103.0 | 28119 | 0.1239 | 0.8164 | 0.7145 | 0.8761 | 0.2994 | 0.0000 |
|
| 160 |
+
| 0.1212 | 104.0 | 28392 | 0.1235 | 0.8171 | 0.7189 | 0.8781 | 0.3015 | 0.0000 |
|
| 161 |
+
| 0.1205 | 105.0 | 28665 | 0.1234 | 0.8191 | 0.7225 | 0.8824 | 0.3018 | 0.0000 |
|
| 162 |
+
| 0.1205 | 106.0 | 28938 | 0.1240 | 0.8179 | 0.7183 | 0.8820 | 0.3022 | 0.0000 |
|
| 163 |
+
| 0.1209 | 107.0 | 29211 | 0.1234 | 0.8188 | 0.7270 | 0.8822 | 0.3015 | 0.0000 |
|
| 164 |
+
| 0.1209 | 108.0 | 29484 | 0.1234 | 0.8162 | 0.7226 | 0.8762 | 0.2998 | 0.0000 |
|
| 165 |
+
| 0.1201 | 109.0 | 29757 | 0.1236 | 0.8216 | 0.7257 | 0.8868 | 0.3035 | 0.0000 |
|
| 166 |
+
| 0.1205 | 110.0 | 30030 | 0.1237 | 0.8204 | 0.7232 | 0.8842 | 0.3032 | 0.0000 |
|
| 167 |
+
| 0.1205 | 111.0 | 30303 | 0.1232 | 0.8178 | 0.7224 | 0.8771 | 0.3053 | 0.0000 |
|
| 168 |
+
|
| 169 |
+
|
| 170 |
+
### Framework versions
|
| 171 |
+
|
| 172 |
+
- Transformers 4.41.1
|
| 173 |
+
- Pytorch 2.3.0+cu121
|
| 174 |
+
- Datasets 2.19.1
|
| 175 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 111.0,
|
| 3 |
+
"eval_accuracy": 0.3182608695652174,
|
| 4 |
+
"eval_f1_macro": 0.7172549519840059,
|
| 5 |
+
"eval_f1_micro": 0.82173859382071,
|
| 6 |
+
"eval_loss": 0.12350255995988846,
|
| 7 |
+
"eval_roc_auc": 0.8829346301484131,
|
| 8 |
+
"eval_runtime": 536.2665,
|
| 9 |
+
"eval_samples_per_second": 5.361,
|
| 10 |
+
"eval_steps_per_second": 0.168,
|
| 11 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 12 |
+
"total_flos": 1.4332659431723232e+21,
|
| 13 |
+
"train_loss": 0.13822040822849538,
|
| 14 |
+
"train_runtime": 237956.1018,
|
| 15 |
+
"train_samples_per_second": 5.489,
|
| 16 |
+
"train_steps_per_second": 0.172
|
| 17 |
+
}
|
logs/events.out.tfevents.1717167376.datavisu2
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3521d09317b6177abdf24c5baff013d1b2a7e781d075ec611ce1b1da654455b
|
| 3 |
+
size 80413
|
logs/events.out.tfevents.1717405982.datavisu2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1dc2c7ab38af042ddaa051b1f78f0aef45743939543eb73aa492c23526d357b4
|
| 3 |
+
size 40
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1228107508
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a268f9568d0716a90b0307ce0efa598acdfa84c6432d1973306680941b129a08
|
| 3 |
size 1228107508
|
test_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 111.0,
|
| 3 |
+
"eval_accuracy": 0.3182608695652174,
|
| 4 |
+
"eval_f1_macro": 0.7172549519840059,
|
| 5 |
+
"eval_f1_micro": 0.82173859382071,
|
| 6 |
+
"eval_loss": 0.12350255995988846,
|
| 7 |
+
"eval_roc_auc": 0.8829346301484131,
|
| 8 |
+
"eval_runtime": 536.2665,
|
| 9 |
+
"eval_samples_per_second": 5.361,
|
| 10 |
+
"eval_steps_per_second": 0.168,
|
| 11 |
+
"learning_rate": 1.0000000000000002e-07
|
| 12 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 111.0,
|
| 3 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 4 |
+
"total_flos": 1.4332659431723232e+21,
|
| 5 |
+
"train_loss": 0.13822040822849538,
|
| 6 |
+
"train_runtime": 237956.1018,
|
| 7 |
+
"train_samples_per_second": 5.489,
|
| 8 |
+
"train_steps_per_second": 0.172
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1915 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.12300820648670197,
|
| 3 |
+
"best_model_checkpoint": "/home1/datawork/mcontini/models/multilabel/huggingface/DinoVdeau-large-2024_05_31-batch-size32_epochs150_freeze/checkpoint-27573",
|
| 4 |
+
"epoch": 111.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 30303,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.23719491234101064,
|
| 14 |
+
"eval_f1_macro": 0.51050515939363,
|
| 15 |
+
"eval_f1_micro": 0.7471888818698673,
|
| 16 |
+
"eval_loss": 0.17266152799129486,
|
| 17 |
+
"eval_roc_auc": 0.8326243241283786,
|
| 18 |
+
"eval_runtime": 528.4205,
|
| 19 |
+
"eval_samples_per_second": 5.505,
|
| 20 |
+
"eval_steps_per_second": 0.172,
|
| 21 |
+
"learning_rate": 0.001,
|
| 22 |
+
"step": 273
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 1.8315018315018317,
|
| 26 |
+
"grad_norm": 0.2667010724544525,
|
| 27 |
+
"learning_rate": 0.001,
|
| 28 |
+
"loss": 0.2751,
|
| 29 |
+
"step": 500
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 2.0,
|
| 33 |
+
"eval_accuracy": 0.2506015812994156,
|
| 34 |
+
"eval_f1_macro": 0.5754792679279648,
|
| 35 |
+
"eval_f1_micro": 0.759063829787234,
|
| 36 |
+
"eval_loss": 0.15843337774276733,
|
| 37 |
+
"eval_roc_auc": 0.8363896810255705,
|
| 38 |
+
"eval_runtime": 540.7289,
|
| 39 |
+
"eval_samples_per_second": 5.38,
|
| 40 |
+
"eval_steps_per_second": 0.168,
|
| 41 |
+
"learning_rate": 0.001,
|
| 42 |
+
"step": 546
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.0,
|
| 46 |
+
"eval_accuracy": 0.24682021313166036,
|
| 47 |
+
"eval_f1_macro": 0.5997966166279454,
|
| 48 |
+
"eval_f1_micro": 0.7712364371467471,
|
| 49 |
+
"eval_loss": 0.15126323699951172,
|
| 50 |
+
"eval_roc_auc": 0.8510669293802386,
|
| 51 |
+
"eval_runtime": 526.2016,
|
| 52 |
+
"eval_samples_per_second": 5.528,
|
| 53 |
+
"eval_steps_per_second": 0.173,
|
| 54 |
+
"learning_rate": 0.001,
|
| 55 |
+
"step": 819
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 3.663003663003663,
|
| 59 |
+
"grad_norm": 0.2700529992580414,
|
| 60 |
+
"learning_rate": 0.001,
|
| 61 |
+
"loss": 0.1714,
|
| 62 |
+
"step": 1000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 4.0,
|
| 66 |
+
"eval_accuracy": 0.2516328635269852,
|
| 67 |
+
"eval_f1_macro": 0.6102972679327003,
|
| 68 |
+
"eval_f1_micro": 0.7649153278332611,
|
| 69 |
+
"eval_loss": 0.14971894025802612,
|
| 70 |
+
"eval_roc_auc": 0.8343823287511987,
|
| 71 |
+
"eval_runtime": 526.9869,
|
| 72 |
+
"eval_samples_per_second": 5.52,
|
| 73 |
+
"eval_steps_per_second": 0.173,
|
| 74 |
+
"learning_rate": 0.001,
|
| 75 |
+
"step": 1092
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.0,
|
| 79 |
+
"eval_accuracy": 0.26194568580268135,
|
| 80 |
+
"eval_f1_macro": 0.6048101972148001,
|
| 81 |
+
"eval_f1_micro": 0.7743272938347361,
|
| 82 |
+
"eval_loss": 0.14786843955516815,
|
| 83 |
+
"eval_roc_auc": 0.851433496224377,
|
| 84 |
+
"eval_runtime": 527.9687,
|
| 85 |
+
"eval_samples_per_second": 5.51,
|
| 86 |
+
"eval_steps_per_second": 0.172,
|
| 87 |
+
"learning_rate": 0.001,
|
| 88 |
+
"step": 1365
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.4945054945054945,
|
| 92 |
+
"grad_norm": 0.22031261026859283,
|
| 93 |
+
"learning_rate": 0.001,
|
| 94 |
+
"loss": 0.1664,
|
| 95 |
+
"step": 1500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 6.0,
|
| 99 |
+
"eval_accuracy": 0.23856995531110348,
|
| 100 |
+
"eval_f1_macro": 0.6142379178757014,
|
| 101 |
+
"eval_f1_micro": 0.7744215397559949,
|
| 102 |
+
"eval_loss": 0.1464792788028717,
|
| 103 |
+
"eval_roc_auc": 0.8475986141620167,
|
| 104 |
+
"eval_runtime": 524.7924,
|
| 105 |
+
"eval_samples_per_second": 5.543,
|
| 106 |
+
"eval_steps_per_second": 0.173,
|
| 107 |
+
"learning_rate": 0.001,
|
| 108 |
+
"step": 1638
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 7.0,
|
| 112 |
+
"eval_accuracy": 0.25232038501203163,
|
| 113 |
+
"eval_f1_macro": 0.6378919485445709,
|
| 114 |
+
"eval_f1_micro": 0.7745991019884542,
|
| 115 |
+
"eval_loss": 0.1471046507358551,
|
| 116 |
+
"eval_roc_auc": 0.8434055352108081,
|
| 117 |
+
"eval_runtime": 526.2444,
|
| 118 |
+
"eval_samples_per_second": 5.528,
|
| 119 |
+
"eval_steps_per_second": 0.173,
|
| 120 |
+
"learning_rate": 0.001,
|
| 121 |
+
"step": 1911
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 7.326007326007326,
|
| 125 |
+
"grad_norm": 0.22752803564071655,
|
| 126 |
+
"learning_rate": 0.001,
|
| 127 |
+
"loss": 0.1613,
|
| 128 |
+
"step": 2000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"eval_accuracy": 0.2492265383293228,
|
| 133 |
+
"eval_f1_macro": 0.6349994110674028,
|
| 134 |
+
"eval_f1_micro": 0.7730045646516785,
|
| 135 |
+
"eval_loss": 0.14884118735790253,
|
| 136 |
+
"eval_roc_auc": 0.8431799269621351,
|
| 137 |
+
"eval_runtime": 522.6769,
|
| 138 |
+
"eval_samples_per_second": 5.566,
|
| 139 |
+
"eval_steps_per_second": 0.174,
|
| 140 |
+
"learning_rate": 0.001,
|
| 141 |
+
"step": 2184
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 9.0,
|
| 145 |
+
"eval_accuracy": 0.17084908903403231,
|
| 146 |
+
"eval_f1_macro": 0.5913962348409779,
|
| 147 |
+
"eval_f1_micro": 0.7571320373643019,
|
| 148 |
+
"eval_loss": 0.1696147322654724,
|
| 149 |
+
"eval_roc_auc": 0.8382156770442014,
|
| 150 |
+
"eval_runtime": 524.2587,
|
| 151 |
+
"eval_samples_per_second": 5.549,
|
| 152 |
+
"eval_steps_per_second": 0.174,
|
| 153 |
+
"learning_rate": 0.001,
|
| 154 |
+
"step": 2457
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 9.157509157509157,
|
| 158 |
+
"grad_norm": 0.1698230803012848,
|
| 159 |
+
"learning_rate": 0.001,
|
| 160 |
+
"loss": 0.1607,
|
| 161 |
+
"step": 2500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 10.0,
|
| 165 |
+
"eval_accuracy": 0.2633207287727741,
|
| 166 |
+
"eval_f1_macro": 0.6155340436417148,
|
| 167 |
+
"eval_f1_micro": 0.7752799457074991,
|
| 168 |
+
"eval_loss": 0.14972557127475739,
|
| 169 |
+
"eval_roc_auc": 0.8458981216058233,
|
| 170 |
+
"eval_runtime": 525.2478,
|
| 171 |
+
"eval_samples_per_second": 5.538,
|
| 172 |
+
"eval_steps_per_second": 0.173,
|
| 173 |
+
"learning_rate": 0.001,
|
| 174 |
+
"step": 2730
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 10.989010989010989,
|
| 178 |
+
"grad_norm": 0.18454667925834656,
|
| 179 |
+
"learning_rate": 0.001,
|
| 180 |
+
"loss": 0.1589,
|
| 181 |
+
"step": 3000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 11.0,
|
| 185 |
+
"eval_accuracy": 0.2591955998624957,
|
| 186 |
+
"eval_f1_macro": 0.6157336320155493,
|
| 187 |
+
"eval_f1_micro": 0.7746945972041475,
|
| 188 |
+
"eval_loss": 0.14687682688236237,
|
| 189 |
+
"eval_roc_auc": 0.8483143864682651,
|
| 190 |
+
"eval_runtime": 527.1812,
|
| 191 |
+
"eval_samples_per_second": 5.518,
|
| 192 |
+
"eval_steps_per_second": 0.173,
|
| 193 |
+
"learning_rate": 0.001,
|
| 194 |
+
"step": 3003
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 12.0,
|
| 198 |
+
"eval_accuracy": 0.25575799243726366,
|
| 199 |
+
"eval_f1_macro": 0.6488857718028787,
|
| 200 |
+
"eval_f1_micro": 0.7830250450155508,
|
| 201 |
+
"eval_loss": 0.14556235074996948,
|
| 202 |
+
"eval_roc_auc": 0.8597932614995001,
|
| 203 |
+
"eval_runtime": 527.5141,
|
| 204 |
+
"eval_samples_per_second": 5.515,
|
| 205 |
+
"eval_steps_per_second": 0.173,
|
| 206 |
+
"learning_rate": 0.001,
|
| 207 |
+
"step": 3276
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 12.820512820512821,
|
| 211 |
+
"grad_norm": 0.16842660307884216,
|
| 212 |
+
"learning_rate": 0.001,
|
| 213 |
+
"loss": 0.1578,
|
| 214 |
+
"step": 3500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 13.0,
|
| 218 |
+
"eval_accuracy": 0.2554142316947405,
|
| 219 |
+
"eval_f1_macro": 0.6580664808969154,
|
| 220 |
+
"eval_f1_micro": 0.7843104596935376,
|
| 221 |
+
"eval_loss": 0.1428324580192566,
|
| 222 |
+
"eval_roc_auc": 0.8557355060660653,
|
| 223 |
+
"eval_runtime": 526.9892,
|
| 224 |
+
"eval_samples_per_second": 5.52,
|
| 225 |
+
"eval_steps_per_second": 0.173,
|
| 226 |
+
"learning_rate": 0.001,
|
| 227 |
+
"step": 3549
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 14.0,
|
| 231 |
+
"eval_accuracy": 0.2506015812994156,
|
| 232 |
+
"eval_f1_macro": 0.6418882151192971,
|
| 233 |
+
"eval_f1_micro": 0.7831899072604227,
|
| 234 |
+
"eval_loss": 0.14352969825267792,
|
| 235 |
+
"eval_roc_auc": 0.8542397387049757,
|
| 236 |
+
"eval_runtime": 552.828,
|
| 237 |
+
"eval_samples_per_second": 5.262,
|
| 238 |
+
"eval_steps_per_second": 0.165,
|
| 239 |
+
"learning_rate": 0.001,
|
| 240 |
+
"step": 3822
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 14.652014652014651,
|
| 244 |
+
"grad_norm": 0.16212721168994904,
|
| 245 |
+
"learning_rate": 0.001,
|
| 246 |
+
"loss": 0.158,
|
| 247 |
+
"step": 4000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 15.0,
|
| 251 |
+
"eval_accuracy": 0.25335166723960123,
|
| 252 |
+
"eval_f1_macro": 0.6573171730380419,
|
| 253 |
+
"eval_f1_micro": 0.7810577597952191,
|
| 254 |
+
"eval_loss": 0.14535032212734222,
|
| 255 |
+
"eval_roc_auc": 0.8562740234249071,
|
| 256 |
+
"eval_runtime": 543.9143,
|
| 257 |
+
"eval_samples_per_second": 5.348,
|
| 258 |
+
"eval_steps_per_second": 0.167,
|
| 259 |
+
"learning_rate": 0.001,
|
| 260 |
+
"step": 4095
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 16.0,
|
| 264 |
+
"eval_accuracy": 0.26400825025782054,
|
| 265 |
+
"eval_f1_macro": 0.6503641322273297,
|
| 266 |
+
"eval_f1_micro": 0.7865547601415288,
|
| 267 |
+
"eval_loss": 0.1448182314634323,
|
| 268 |
+
"eval_roc_auc": 0.8602699185077041,
|
| 269 |
+
"eval_runtime": 528.0191,
|
| 270 |
+
"eval_samples_per_second": 5.509,
|
| 271 |
+
"eval_steps_per_second": 0.172,
|
| 272 |
+
"learning_rate": 0.001,
|
| 273 |
+
"step": 4368
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 16.483516483516482,
|
| 277 |
+
"grad_norm": 0.15847235918045044,
|
| 278 |
+
"learning_rate": 0.001,
|
| 279 |
+
"loss": 0.1587,
|
| 280 |
+
"step": 4500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 17.0,
|
| 284 |
+
"eval_accuracy": 0.2595393606050189,
|
| 285 |
+
"eval_f1_macro": 0.6441131277200526,
|
| 286 |
+
"eval_f1_micro": 0.7819386012413047,
|
| 287 |
+
"eval_loss": 0.14306068420410156,
|
| 288 |
+
"eval_roc_auc": 0.8543391801851071,
|
| 289 |
+
"eval_runtime": 533.465,
|
| 290 |
+
"eval_samples_per_second": 5.453,
|
| 291 |
+
"eval_steps_per_second": 0.171,
|
| 292 |
+
"learning_rate": 0.001,
|
| 293 |
+
"step": 4641
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 18.0,
|
| 297 |
+
"eval_accuracy": 0.25472671020969406,
|
| 298 |
+
"eval_f1_macro": 0.6412796509721743,
|
| 299 |
+
"eval_f1_micro": 0.7838684089675614,
|
| 300 |
+
"eval_loss": 0.1446852684020996,
|
| 301 |
+
"eval_roc_auc": 0.8547791265270707,
|
| 302 |
+
"eval_runtime": 537.7058,
|
| 303 |
+
"eval_samples_per_second": 5.41,
|
| 304 |
+
"eval_steps_per_second": 0.169,
|
| 305 |
+
"learning_rate": 0.001,
|
| 306 |
+
"step": 4914
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 18.315018315018314,
|
| 310 |
+
"grad_norm": 0.14501237869262695,
|
| 311 |
+
"learning_rate": 0.001,
|
| 312 |
+
"loss": 0.1571,
|
| 313 |
+
"step": 5000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 19.0,
|
| 317 |
+
"eval_accuracy": 0.26469577174286696,
|
| 318 |
+
"eval_f1_macro": 0.6441719729073186,
|
| 319 |
+
"eval_f1_micro": 0.7836073910832345,
|
| 320 |
+
"eval_loss": 0.1414887011051178,
|
| 321 |
+
"eval_roc_auc": 0.8505868762044602,
|
| 322 |
+
"eval_runtime": 542.1965,
|
| 323 |
+
"eval_samples_per_second": 5.365,
|
| 324 |
+
"eval_steps_per_second": 0.168,
|
| 325 |
+
"learning_rate": 0.001,
|
| 326 |
+
"step": 5187
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 20.0,
|
| 330 |
+
"eval_accuracy": 0.2609144035751117,
|
| 331 |
+
"eval_f1_macro": 0.6436752083994657,
|
| 332 |
+
"eval_f1_micro": 0.7838809251976828,
|
| 333 |
+
"eval_loss": 0.1405678391456604,
|
| 334 |
+
"eval_roc_auc": 0.8513351741993039,
|
| 335 |
+
"eval_runtime": 537.1864,
|
| 336 |
+
"eval_samples_per_second": 5.415,
|
| 337 |
+
"eval_steps_per_second": 0.169,
|
| 338 |
+
"learning_rate": 0.001,
|
| 339 |
+
"step": 5460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 20.146520146520146,
|
| 343 |
+
"grad_norm": 0.16518211364746094,
|
| 344 |
+
"learning_rate": 0.001,
|
| 345 |
+
"loss": 0.1577,
|
| 346 |
+
"step": 5500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 21.0,
|
| 350 |
+
"eval_accuracy": 0.25747679614987967,
|
| 351 |
+
"eval_f1_macro": 0.6461144436259285,
|
| 352 |
+
"eval_f1_micro": 0.7847450484618627,
|
| 353 |
+
"eval_loss": 0.14136537909507751,
|
| 354 |
+
"eval_roc_auc": 0.8527206442391784,
|
| 355 |
+
"eval_runtime": 535.7541,
|
| 356 |
+
"eval_samples_per_second": 5.43,
|
| 357 |
+
"eval_steps_per_second": 0.17,
|
| 358 |
+
"learning_rate": 0.001,
|
| 359 |
+
"step": 5733
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 21.978021978021978,
|
| 363 |
+
"grad_norm": 0.14056652784347534,
|
| 364 |
+
"learning_rate": 0.001,
|
| 365 |
+
"loss": 0.1556,
|
| 366 |
+
"step": 6000
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 22.0,
|
| 370 |
+
"eval_accuracy": 0.26194568580268135,
|
| 371 |
+
"eval_f1_macro": 0.6587430664519773,
|
| 372 |
+
"eval_f1_micro": 0.78838833814364,
|
| 373 |
+
"eval_loss": 0.1386082023382187,
|
| 374 |
+
"eval_roc_auc": 0.8567127686586293,
|
| 375 |
+
"eval_runtime": 534.4718,
|
| 376 |
+
"eval_samples_per_second": 5.443,
|
| 377 |
+
"eval_steps_per_second": 0.17,
|
| 378 |
+
"learning_rate": 0.001,
|
| 379 |
+
"step": 6006
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 23.0,
|
| 383 |
+
"eval_accuracy": 0.268133379168099,
|
| 384 |
+
"eval_f1_macro": 0.6502835561459052,
|
| 385 |
+
"eval_f1_micro": 0.7900436534586972,
|
| 386 |
+
"eval_loss": 0.13822348415851593,
|
| 387 |
+
"eval_roc_auc": 0.8566583659653727,
|
| 388 |
+
"eval_runtime": 532.9053,
|
| 389 |
+
"eval_samples_per_second": 5.459,
|
| 390 |
+
"eval_steps_per_second": 0.171,
|
| 391 |
+
"learning_rate": 0.001,
|
| 392 |
+
"step": 6279
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 23.80952380952381,
|
| 396 |
+
"grad_norm": 0.19816666841506958,
|
| 397 |
+
"learning_rate": 0.001,
|
| 398 |
+
"loss": 0.1558,
|
| 399 |
+
"step": 6500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 24.0,
|
| 403 |
+
"eval_accuracy": 0.2650395324853902,
|
| 404 |
+
"eval_f1_macro": 0.6519157870757598,
|
| 405 |
+
"eval_f1_micro": 0.7880997276346112,
|
| 406 |
+
"eval_loss": 0.13893215358257294,
|
| 407 |
+
"eval_roc_auc": 0.8560751822992996,
|
| 408 |
+
"eval_runtime": 537.3622,
|
| 409 |
+
"eval_samples_per_second": 5.413,
|
| 410 |
+
"eval_steps_per_second": 0.169,
|
| 411 |
+
"learning_rate": 0.001,
|
| 412 |
+
"step": 6552
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 25.0,
|
| 416 |
+
"eval_accuracy": 0.2695084221381918,
|
| 417 |
+
"eval_f1_macro": 0.6553890148296274,
|
| 418 |
+
"eval_f1_micro": 0.7883266848624816,
|
| 419 |
+
"eval_loss": 0.1396929770708084,
|
| 420 |
+
"eval_roc_auc": 0.8556256648938046,
|
| 421 |
+
"eval_runtime": 530.4438,
|
| 422 |
+
"eval_samples_per_second": 5.484,
|
| 423 |
+
"eval_steps_per_second": 0.172,
|
| 424 |
+
"learning_rate": 0.001,
|
| 425 |
+
"step": 6825
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 25.641025641025642,
|
| 429 |
+
"grad_norm": 0.160660982131958,
|
| 430 |
+
"learning_rate": 0.001,
|
| 431 |
+
"loss": 0.1551,
|
| 432 |
+
"step": 7000
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 26.0,
|
| 436 |
+
"eval_accuracy": 0.2650395324853902,
|
| 437 |
+
"eval_f1_macro": 0.6349973216617472,
|
| 438 |
+
"eval_f1_micro": 0.7793733121865489,
|
| 439 |
+
"eval_loss": 0.14012907445430756,
|
| 440 |
+
"eval_roc_auc": 0.8453129082029109,
|
| 441 |
+
"eval_runtime": 532.4751,
|
| 442 |
+
"eval_samples_per_second": 5.463,
|
| 443 |
+
"eval_steps_per_second": 0.171,
|
| 444 |
+
"learning_rate": 0.001,
|
| 445 |
+
"step": 7098
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 27.0,
|
| 449 |
+
"eval_accuracy": 0.2499140598143692,
|
| 450 |
+
"eval_f1_macro": 0.6667665130743651,
|
| 451 |
+
"eval_f1_micro": 0.7915316128502852,
|
| 452 |
+
"eval_loss": 0.1402139514684677,
|
| 453 |
+
"eval_roc_auc": 0.8637912699884471,
|
| 454 |
+
"eval_runtime": 535.9247,
|
| 455 |
+
"eval_samples_per_second": 5.428,
|
| 456 |
+
"eval_steps_per_second": 0.17,
|
| 457 |
+
"learning_rate": 0.001,
|
| 458 |
+
"step": 7371
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 27.47252747252747,
|
| 462 |
+
"grad_norm": 0.15610536932945251,
|
| 463 |
+
"learning_rate": 0.001,
|
| 464 |
+
"loss": 0.156,
|
| 465 |
+
"step": 7500
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 28.0,
|
| 469 |
+
"eval_accuracy": 0.26400825025782054,
|
| 470 |
+
"eval_f1_macro": 0.6575889205208383,
|
| 471 |
+
"eval_f1_micro": 0.7907676869041647,
|
| 472 |
+
"eval_loss": 0.1389472633600235,
|
| 473 |
+
"eval_roc_auc": 0.8581090945395875,
|
| 474 |
+
"eval_runtime": 536.3338,
|
| 475 |
+
"eval_samples_per_second": 5.424,
|
| 476 |
+
"eval_steps_per_second": 0.17,
|
| 477 |
+
"learning_rate": 0.001,
|
| 478 |
+
"step": 7644
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 29.0,
|
| 482 |
+
"eval_accuracy": 0.268133379168099,
|
| 483 |
+
"eval_f1_macro": 0.6657248014277785,
|
| 484 |
+
"eval_f1_micro": 0.793982620101656,
|
| 485 |
+
"eval_loss": 0.14010308682918549,
|
| 486 |
+
"eval_roc_auc": 0.8654152756895732,
|
| 487 |
+
"eval_runtime": 547.3885,
|
| 488 |
+
"eval_samples_per_second": 5.314,
|
| 489 |
+
"eval_steps_per_second": 0.166,
|
| 490 |
+
"learning_rate": 0.001,
|
| 491 |
+
"step": 7917
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 29.304029304029303,
|
| 495 |
+
"grad_norm": 0.15964308381080627,
|
| 496 |
+
"learning_rate": 0.0001,
|
| 497 |
+
"loss": 0.155,
|
| 498 |
+
"step": 8000
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 30.0,
|
| 502 |
+
"eval_accuracy": 0.2739773117909935,
|
| 503 |
+
"eval_f1_macro": 0.6710788331447379,
|
| 504 |
+
"eval_f1_micro": 0.7986502613061192,
|
| 505 |
+
"eval_loss": 0.13457615673542023,
|
| 506 |
+
"eval_roc_auc": 0.8668929367093478,
|
| 507 |
+
"eval_runtime": 547.6107,
|
| 508 |
+
"eval_samples_per_second": 5.312,
|
| 509 |
+
"eval_steps_per_second": 0.166,
|
| 510 |
+
"learning_rate": 0.0001,
|
| 511 |
+
"step": 8190
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 31.0,
|
| 515 |
+
"eval_accuracy": 0.2784462014437951,
|
| 516 |
+
"eval_f1_macro": 0.6773801465956272,
|
| 517 |
+
"eval_f1_micro": 0.7977876476996564,
|
| 518 |
+
"eval_loss": 0.1323619782924652,
|
| 519 |
+
"eval_roc_auc": 0.8613412783291686,
|
| 520 |
+
"eval_runtime": 550.0815,
|
| 521 |
+
"eval_samples_per_second": 5.288,
|
| 522 |
+
"eval_steps_per_second": 0.165,
|
| 523 |
+
"learning_rate": 0.0001,
|
| 524 |
+
"step": 8463
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 31.135531135531135,
|
| 528 |
+
"grad_norm": 0.15679965913295746,
|
| 529 |
+
"learning_rate": 0.0001,
|
| 530 |
+
"loss": 0.1466,
|
| 531 |
+
"step": 8500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 32.0,
|
| 535 |
+
"eval_accuracy": 0.27294602956342384,
|
| 536 |
+
"eval_f1_macro": 0.6778747090268944,
|
| 537 |
+
"eval_f1_micro": 0.8018979833926453,
|
| 538 |
+
"eval_loss": 0.13328427076339722,
|
| 539 |
+
"eval_roc_auc": 0.8704036395061567,
|
| 540 |
+
"eval_runtime": 536.2816,
|
| 541 |
+
"eval_samples_per_second": 5.424,
|
| 542 |
+
"eval_steps_per_second": 0.17,
|
| 543 |
+
"learning_rate": 0.0001,
|
| 544 |
+
"step": 8736
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 32.967032967032964,
|
| 548 |
+
"grad_norm": 0.16012457013130188,
|
| 549 |
+
"learning_rate": 0.0001,
|
| 550 |
+
"loss": 0.1431,
|
| 551 |
+
"step": 9000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 33.0,
|
| 555 |
+
"eval_accuracy": 0.2794774836713647,
|
| 556 |
+
"eval_f1_macro": 0.6849041252254977,
|
| 557 |
+
"eval_f1_micro": 0.8033507506013103,
|
| 558 |
+
"eval_loss": 0.13171622157096863,
|
| 559 |
+
"eval_roc_auc": 0.8672834939642976,
|
| 560 |
+
"eval_runtime": 523.1156,
|
| 561 |
+
"eval_samples_per_second": 5.561,
|
| 562 |
+
"eval_steps_per_second": 0.174,
|
| 563 |
+
"learning_rate": 0.0001,
|
| 564 |
+
"step": 9009
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 34.0,
|
| 568 |
+
"eval_accuracy": 0.27982124441388795,
|
| 569 |
+
"eval_f1_macro": 0.6836107081350358,
|
| 570 |
+
"eval_f1_micro": 0.8020612558700079,
|
| 571 |
+
"eval_loss": 0.1307307630777359,
|
| 572 |
+
"eval_roc_auc": 0.8659768313427901,
|
| 573 |
+
"eval_runtime": 537.0527,
|
| 574 |
+
"eval_samples_per_second": 5.417,
|
| 575 |
+
"eval_steps_per_second": 0.169,
|
| 576 |
+
"learning_rate": 0.0001,
|
| 577 |
+
"step": 9282
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 34.798534798534796,
|
| 581 |
+
"grad_norm": 0.1809278130531311,
|
| 582 |
+
"learning_rate": 0.0001,
|
| 583 |
+
"loss": 0.1413,
|
| 584 |
+
"step": 9500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 35.0,
|
| 588 |
+
"eval_accuracy": 0.2853214162942592,
|
| 589 |
+
"eval_f1_macro": 0.6903143910655651,
|
| 590 |
+
"eval_f1_micro": 0.8046306144154465,
|
| 591 |
+
"eval_loss": 0.13025963306427002,
|
| 592 |
+
"eval_roc_auc": 0.8719100555511805,
|
| 593 |
+
"eval_runtime": 531.2738,
|
| 594 |
+
"eval_samples_per_second": 5.476,
|
| 595 |
+
"eval_steps_per_second": 0.171,
|
| 596 |
+
"learning_rate": 0.0001,
|
| 597 |
+
"step": 9555
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 36.0,
|
| 601 |
+
"eval_accuracy": 0.290134066689584,
|
| 602 |
+
"eval_f1_macro": 0.6931792274102746,
|
| 603 |
+
"eval_f1_micro": 0.8050928824879983,
|
| 604 |
+
"eval_loss": 0.1296597272157669,
|
| 605 |
+
"eval_roc_auc": 0.8663567399138056,
|
| 606 |
+
"eval_runtime": 536.3308,
|
| 607 |
+
"eval_samples_per_second": 5.424,
|
| 608 |
+
"eval_steps_per_second": 0.17,
|
| 609 |
+
"learning_rate": 0.0001,
|
| 610 |
+
"step": 9828
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 36.63003663003663,
|
| 614 |
+
"grad_norm": 0.19757141172885895,
|
| 615 |
+
"learning_rate": 0.0001,
|
| 616 |
+
"loss": 0.14,
|
| 617 |
+
"step": 10000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 37.0,
|
| 621 |
+
"eval_accuracy": 0.288415262976968,
|
| 622 |
+
"eval_f1_macro": 0.6884731439645044,
|
| 623 |
+
"eval_f1_micro": 0.8034535718733136,
|
| 624 |
+
"eval_loss": 0.12961770594120026,
|
| 625 |
+
"eval_roc_auc": 0.8670685343995587,
|
| 626 |
+
"eval_runtime": 530.0276,
|
| 627 |
+
"eval_samples_per_second": 5.488,
|
| 628 |
+
"eval_steps_per_second": 0.172,
|
| 629 |
+
"learning_rate": 0.0001,
|
| 630 |
+
"step": 10101
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 38.0,
|
| 634 |
+
"eval_accuracy": 0.28704022000687524,
|
| 635 |
+
"eval_f1_macro": 0.68878596648098,
|
| 636 |
+
"eval_f1_micro": 0.8042539049518111,
|
| 637 |
+
"eval_loss": 0.12927678227424622,
|
| 638 |
+
"eval_roc_auc": 0.8672822510214007,
|
| 639 |
+
"eval_runtime": 524.5751,
|
| 640 |
+
"eval_samples_per_second": 5.545,
|
| 641 |
+
"eval_steps_per_second": 0.173,
|
| 642 |
+
"learning_rate": 0.0001,
|
| 643 |
+
"step": 10374
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 38.46153846153846,
|
| 647 |
+
"grad_norm": 0.16473665833473206,
|
| 648 |
+
"learning_rate": 0.0001,
|
| 649 |
+
"loss": 0.1391,
|
| 650 |
+
"step": 10500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 39.0,
|
| 654 |
+
"eval_accuracy": 0.2877277414919216,
|
| 655 |
+
"eval_f1_macro": 0.6869571453655325,
|
| 656 |
+
"eval_f1_micro": 0.8032656478961692,
|
| 657 |
+
"eval_loss": 0.13043531775474548,
|
| 658 |
+
"eval_roc_auc": 0.8645365442953599,
|
| 659 |
+
"eval_runtime": 528.3251,
|
| 660 |
+
"eval_samples_per_second": 5.506,
|
| 661 |
+
"eval_steps_per_second": 0.172,
|
| 662 |
+
"learning_rate": 0.0001,
|
| 663 |
+
"step": 10647
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 40.0,
|
| 667 |
+
"eval_accuracy": 0.29082158817463044,
|
| 668 |
+
"eval_f1_macro": 0.693512791025367,
|
| 669 |
+
"eval_f1_micro": 0.8063839414256921,
|
| 670 |
+
"eval_loss": 0.12887024879455566,
|
| 671 |
+
"eval_roc_auc": 0.871282703561486,
|
| 672 |
+
"eval_runtime": 528.794,
|
| 673 |
+
"eval_samples_per_second": 5.501,
|
| 674 |
+
"eval_steps_per_second": 0.172,
|
| 675 |
+
"learning_rate": 0.0001,
|
| 676 |
+
"step": 10920
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 40.29304029304029,
|
| 680 |
+
"grad_norm": 0.20218676328659058,
|
| 681 |
+
"learning_rate": 0.0001,
|
| 682 |
+
"loss": 0.1375,
|
| 683 |
+
"step": 11000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 41.0,
|
| 687 |
+
"eval_accuracy": 0.28704022000687524,
|
| 688 |
+
"eval_f1_macro": 0.6924207618698955,
|
| 689 |
+
"eval_f1_micro": 0.8037148594377511,
|
| 690 |
+
"eval_loss": 0.12975196540355682,
|
| 691 |
+
"eval_roc_auc": 0.8650046241911243,
|
| 692 |
+
"eval_runtime": 527.8939,
|
| 693 |
+
"eval_samples_per_second": 5.511,
|
| 694 |
+
"eval_steps_per_second": 0.172,
|
| 695 |
+
"learning_rate": 0.0001,
|
| 696 |
+
"step": 11193
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 42.0,
|
| 700 |
+
"eval_accuracy": 0.29597799931247853,
|
| 701 |
+
"eval_f1_macro": 0.6899593151302184,
|
| 702 |
+
"eval_f1_micro": 0.8086140163056905,
|
| 703 |
+
"eval_loss": 0.12854912877082825,
|
| 704 |
+
"eval_roc_auc": 0.8722169989386699,
|
| 705 |
+
"eval_runtime": 524.3169,
|
| 706 |
+
"eval_samples_per_second": 5.548,
|
| 707 |
+
"eval_steps_per_second": 0.174,
|
| 708 |
+
"learning_rate": 0.0001,
|
| 709 |
+
"step": 11466
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 42.124542124542124,
|
| 713 |
+
"grad_norm": 0.16020864248275757,
|
| 714 |
+
"learning_rate": 0.0001,
|
| 715 |
+
"loss": 0.1369,
|
| 716 |
+
"step": 11500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 43.0,
|
| 720 |
+
"eval_accuracy": 0.28704022000687524,
|
| 721 |
+
"eval_f1_macro": 0.6897522012507461,
|
| 722 |
+
"eval_f1_micro": 0.8066790352504639,
|
| 723 |
+
"eval_loss": 0.12848526239395142,
|
| 724 |
+
"eval_roc_auc": 0.8707810621913787,
|
| 725 |
+
"eval_runtime": 525.3547,
|
| 726 |
+
"eval_samples_per_second": 5.537,
|
| 727 |
+
"eval_steps_per_second": 0.173,
|
| 728 |
+
"learning_rate": 0.0001,
|
| 729 |
+
"step": 11739
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 43.956043956043956,
|
| 733 |
+
"grad_norm": 0.19073694944381714,
|
| 734 |
+
"learning_rate": 0.0001,
|
| 735 |
+
"loss": 0.1356,
|
| 736 |
+
"step": 12000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 44.0,
|
| 740 |
+
"eval_accuracy": 0.2921966311447233,
|
| 741 |
+
"eval_f1_macro": 0.6930773017154849,
|
| 742 |
+
"eval_f1_micro": 0.8077876984126985,
|
| 743 |
+
"eval_loss": 0.12860073149204254,
|
| 744 |
+
"eval_roc_auc": 0.8706428384735121,
|
| 745 |
+
"eval_runtime": 525.2422,
|
| 746 |
+
"eval_samples_per_second": 5.538,
|
| 747 |
+
"eval_steps_per_second": 0.173,
|
| 748 |
+
"learning_rate": 0.0001,
|
| 749 |
+
"step": 12012
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 45.0,
|
| 753 |
+
"eval_accuracy": 0.2956342385699553,
|
| 754 |
+
"eval_f1_macro": 0.7036740023156066,
|
| 755 |
+
"eval_f1_micro": 0.8106891471599279,
|
| 756 |
+
"eval_loss": 0.12759028375148773,
|
| 757 |
+
"eval_roc_auc": 0.875156409696237,
|
| 758 |
+
"eval_runtime": 533.8409,
|
| 759 |
+
"eval_samples_per_second": 5.449,
|
| 760 |
+
"eval_steps_per_second": 0.17,
|
| 761 |
+
"learning_rate": 0.0001,
|
| 762 |
+
"step": 12285
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 45.78754578754579,
|
| 766 |
+
"grad_norm": 0.21702325344085693,
|
| 767 |
+
"learning_rate": 0.0001,
|
| 768 |
+
"loss": 0.1348,
|
| 769 |
+
"step": 12500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 46.0,
|
| 773 |
+
"eval_accuracy": 0.2956342385699553,
|
| 774 |
+
"eval_f1_macro": 0.6950811210030032,
|
| 775 |
+
"eval_f1_micro": 0.8078483318155477,
|
| 776 |
+
"eval_loss": 0.12775476276874542,
|
| 777 |
+
"eval_roc_auc": 0.8702717141720853,
|
| 778 |
+
"eval_runtime": 527.4738,
|
| 779 |
+
"eval_samples_per_second": 5.515,
|
| 780 |
+
"eval_steps_per_second": 0.173,
|
| 781 |
+
"learning_rate": 0.0001,
|
| 782 |
+
"step": 12558
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 47.0,
|
| 786 |
+
"eval_accuracy": 0.2918528704022001,
|
| 787 |
+
"eval_f1_macro": 0.7062619936159387,
|
| 788 |
+
"eval_f1_micro": 0.811402081977879,
|
| 789 |
+
"eval_loss": 0.12771955132484436,
|
| 790 |
+
"eval_roc_auc": 0.8774206947996357,
|
| 791 |
+
"eval_runtime": 527.1794,
|
| 792 |
+
"eval_samples_per_second": 5.518,
|
| 793 |
+
"eval_steps_per_second": 0.173,
|
| 794 |
+
"learning_rate": 0.0001,
|
| 795 |
+
"step": 12831
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 47.61904761904762,
|
| 799 |
+
"grad_norm": 0.24386179447174072,
|
| 800 |
+
"learning_rate": 0.0001,
|
| 801 |
+
"loss": 0.1343,
|
| 802 |
+
"step": 13000
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 48.0,
|
| 806 |
+
"eval_accuracy": 0.29082158817463044,
|
| 807 |
+
"eval_f1_macro": 0.6955184040970307,
|
| 808 |
+
"eval_f1_micro": 0.8052384150436536,
|
| 809 |
+
"eval_loss": 0.12763886153697968,
|
| 810 |
+
"eval_roc_auc": 0.8648853772778264,
|
| 811 |
+
"eval_runtime": 528.8318,
|
| 812 |
+
"eval_samples_per_second": 5.501,
|
| 813 |
+
"eval_steps_per_second": 0.172,
|
| 814 |
+
"learning_rate": 0.0001,
|
| 815 |
+
"step": 13104
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 49.0,
|
| 819 |
+
"eval_accuracy": 0.288415262976968,
|
| 820 |
+
"eval_f1_macro": 0.7020626151896329,
|
| 821 |
+
"eval_f1_micro": 0.8098617549329287,
|
| 822 |
+
"eval_loss": 0.12784114480018616,
|
| 823 |
+
"eval_roc_auc": 0.8739923629273905,
|
| 824 |
+
"eval_runtime": 526.9073,
|
| 825 |
+
"eval_samples_per_second": 5.521,
|
| 826 |
+
"eval_steps_per_second": 0.173,
|
| 827 |
+
"learning_rate": 0.0001,
|
| 828 |
+
"step": 13377
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 49.45054945054945,
|
| 832 |
+
"grad_norm": 0.21291495859622955,
|
| 833 |
+
"learning_rate": 0.0001,
|
| 834 |
+
"loss": 0.1344,
|
| 835 |
+
"step": 13500
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 50.0,
|
| 839 |
+
"eval_accuracy": 0.2939154348573393,
|
| 840 |
+
"eval_f1_macro": 0.705975464457343,
|
| 841 |
+
"eval_f1_micro": 0.8101997029212742,
|
| 842 |
+
"eval_loss": 0.12698890268802643,
|
| 843 |
+
"eval_roc_auc": 0.8724947311519554,
|
| 844 |
+
"eval_runtime": 522.9793,
|
| 845 |
+
"eval_samples_per_second": 5.562,
|
| 846 |
+
"eval_steps_per_second": 0.174,
|
| 847 |
+
"learning_rate": 0.0001,
|
| 848 |
+
"step": 13650
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 51.0,
|
| 852 |
+
"eval_accuracy": 0.29769680302509455,
|
| 853 |
+
"eval_f1_macro": 0.709798905916108,
|
| 854 |
+
"eval_f1_micro": 0.8129121550109908,
|
| 855 |
+
"eval_loss": 0.12740205228328705,
|
| 856 |
+
"eval_roc_auc": 0.8779523759725448,
|
| 857 |
+
"eval_runtime": 527.2695,
|
| 858 |
+
"eval_samples_per_second": 5.517,
|
| 859 |
+
"eval_steps_per_second": 0.173,
|
| 860 |
+
"learning_rate": 0.0001,
|
| 861 |
+
"step": 13923
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 51.282051282051285,
|
| 865 |
+
"grad_norm": 0.25488388538360596,
|
| 866 |
+
"learning_rate": 0.0001,
|
| 867 |
+
"loss": 0.1331,
|
| 868 |
+
"step": 14000
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 52.0,
|
| 872 |
+
"eval_accuracy": 0.2918528704022001,
|
| 873 |
+
"eval_f1_macro": 0.707621451538995,
|
| 874 |
+
"eval_f1_micro": 0.8121578560339897,
|
| 875 |
+
"eval_loss": 0.12743453681468964,
|
| 876 |
+
"eval_roc_auc": 0.8764757925731732,
|
| 877 |
+
"eval_runtime": 523.8403,
|
| 878 |
+
"eval_samples_per_second": 5.553,
|
| 879 |
+
"eval_steps_per_second": 0.174,
|
| 880 |
+
"learning_rate": 0.0001,
|
| 881 |
+
"step": 14196
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 53.0,
|
| 885 |
+
"eval_accuracy": 0.29322791337229287,
|
| 886 |
+
"eval_f1_macro": 0.694158242732084,
|
| 887 |
+
"eval_f1_micro": 0.8083623693379792,
|
| 888 |
+
"eval_loss": 0.12782631814479828,
|
| 889 |
+
"eval_roc_auc": 0.8699585560701117,
|
| 890 |
+
"eval_runtime": 524.8463,
|
| 891 |
+
"eval_samples_per_second": 5.543,
|
| 892 |
+
"eval_steps_per_second": 0.173,
|
| 893 |
+
"learning_rate": 0.0001,
|
| 894 |
+
"step": 14469
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 53.11355311355312,
|
| 898 |
+
"grad_norm": 0.2432054579257965,
|
| 899 |
+
"learning_rate": 0.0001,
|
| 900 |
+
"loss": 0.1325,
|
| 901 |
+
"step": 14500
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 54.0,
|
| 905 |
+
"eval_accuracy": 0.2939154348573393,
|
| 906 |
+
"eval_f1_macro": 0.6945407396863954,
|
| 907 |
+
"eval_f1_micro": 0.8095710389288371,
|
| 908 |
+
"eval_loss": 0.12641482055187225,
|
| 909 |
+
"eval_roc_auc": 0.8716949306639478,
|
| 910 |
+
"eval_runtime": 523.4227,
|
| 911 |
+
"eval_samples_per_second": 5.558,
|
| 912 |
+
"eval_steps_per_second": 0.174,
|
| 913 |
+
"learning_rate": 0.0001,
|
| 914 |
+
"step": 14742
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 54.94505494505494,
|
| 918 |
+
"grad_norm": 0.2511921226978302,
|
| 919 |
+
"learning_rate": 0.0001,
|
| 920 |
+
"loss": 0.1318,
|
| 921 |
+
"step": 15000
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 55.0,
|
| 925 |
+
"eval_accuracy": 0.29838432451014096,
|
| 926 |
+
"eval_f1_macro": 0.711364278725223,
|
| 927 |
+
"eval_f1_micro": 0.8117057825241112,
|
| 928 |
+
"eval_loss": 0.12613853812217712,
|
| 929 |
+
"eval_roc_auc": 0.8724015462209637,
|
| 930 |
+
"eval_runtime": 528.7875,
|
| 931 |
+
"eval_samples_per_second": 5.501,
|
| 932 |
+
"eval_steps_per_second": 0.172,
|
| 933 |
+
"learning_rate": 0.0001,
|
| 934 |
+
"step": 15015
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 56.0,
|
| 938 |
+
"eval_accuracy": 0.2990718459951873,
|
| 939 |
+
"eval_f1_macro": 0.7063129571309116,
|
| 940 |
+
"eval_f1_micro": 0.8113456464379947,
|
| 941 |
+
"eval_loss": 0.12689372897148132,
|
| 942 |
+
"eval_roc_auc": 0.8733654252519948,
|
| 943 |
+
"eval_runtime": 525.7416,
|
| 944 |
+
"eval_samples_per_second": 5.533,
|
| 945 |
+
"eval_steps_per_second": 0.173,
|
| 946 |
+
"learning_rate": 0.0001,
|
| 947 |
+
"step": 15288
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 56.776556776556774,
|
| 951 |
+
"grad_norm": 0.305794894695282,
|
| 952 |
+
"learning_rate": 0.0001,
|
| 953 |
+
"loss": 0.1311,
|
| 954 |
+
"step": 15500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 57.0,
|
| 958 |
+
"eval_accuracy": 0.2990718459951873,
|
| 959 |
+
"eval_f1_macro": 0.7092631560861781,
|
| 960 |
+
"eval_f1_micro": 0.8135370461639561,
|
| 961 |
+
"eval_loss": 0.12595032155513763,
|
| 962 |
+
"eval_roc_auc": 0.8756797627485114,
|
| 963 |
+
"eval_runtime": 537.7699,
|
| 964 |
+
"eval_samples_per_second": 5.409,
|
| 965 |
+
"eval_steps_per_second": 0.169,
|
| 966 |
+
"learning_rate": 0.0001,
|
| 967 |
+
"step": 15561
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 58.0,
|
| 971 |
+
"eval_accuracy": 0.29254039188724645,
|
| 972 |
+
"eval_f1_macro": 0.7038026958133716,
|
| 973 |
+
"eval_f1_micro": 0.8129296235679215,
|
| 974 |
+
"eval_loss": 0.12691068649291992,
|
| 975 |
+
"eval_roc_auc": 0.8764486948051197,
|
| 976 |
+
"eval_runtime": 534.4471,
|
| 977 |
+
"eval_samples_per_second": 5.443,
|
| 978 |
+
"eval_steps_per_second": 0.17,
|
| 979 |
+
"learning_rate": 0.0001,
|
| 980 |
+
"step": 15834
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 58.608058608058606,
|
| 984 |
+
"grad_norm": 0.29350945353507996,
|
| 985 |
+
"learning_rate": 0.0001,
|
| 986 |
+
"loss": 0.1307,
|
| 987 |
+
"step": 16000
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 59.0,
|
| 991 |
+
"eval_accuracy": 0.2921966311447233,
|
| 992 |
+
"eval_f1_macro": 0.7074106748802989,
|
| 993 |
+
"eval_f1_micro": 0.8112015199702616,
|
| 994 |
+
"eval_loss": 0.1266162097454071,
|
| 995 |
+
"eval_roc_auc": 0.8727472300115326,
|
| 996 |
+
"eval_runtime": 535.4803,
|
| 997 |
+
"eval_samples_per_second": 5.433,
|
| 998 |
+
"eval_steps_per_second": 0.17,
|
| 999 |
+
"learning_rate": 0.0001,
|
| 1000 |
+
"step": 16107
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 60.0,
|
| 1004 |
+
"eval_accuracy": 0.29322791337229287,
|
| 1005 |
+
"eval_f1_macro": 0.6985455534303677,
|
| 1006 |
+
"eval_f1_micro": 0.8090558527179996,
|
| 1007 |
+
"eval_loss": 0.1263982653617859,
|
| 1008 |
+
"eval_roc_auc": 0.8704458199275943,
|
| 1009 |
+
"eval_runtime": 538.1311,
|
| 1010 |
+
"eval_samples_per_second": 5.406,
|
| 1011 |
+
"eval_steps_per_second": 0.169,
|
| 1012 |
+
"learning_rate": 0.0001,
|
| 1013 |
+
"step": 16380
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 60.43956043956044,
|
| 1017 |
+
"grad_norm": 0.24772068858146667,
|
| 1018 |
+
"learning_rate": 0.0001,
|
| 1019 |
+
"loss": 0.1303,
|
| 1020 |
+
"step": 16500
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 61.0,
|
| 1024 |
+
"eval_accuracy": 0.29597799931247853,
|
| 1025 |
+
"eval_f1_macro": 0.7041363825305129,
|
| 1026 |
+
"eval_f1_micro": 0.8097471110366773,
|
| 1027 |
+
"eval_loss": 0.12581084668636322,
|
| 1028 |
+
"eval_roc_auc": 0.8680428967920301,
|
| 1029 |
+
"eval_runtime": 536.9217,
|
| 1030 |
+
"eval_samples_per_second": 5.418,
|
| 1031 |
+
"eval_steps_per_second": 0.169,
|
| 1032 |
+
"learning_rate": 0.0001,
|
| 1033 |
+
"step": 16653
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 62.0,
|
| 1037 |
+
"eval_accuracy": 0.2911653489171537,
|
| 1038 |
+
"eval_f1_macro": 0.7066021008547532,
|
| 1039 |
+
"eval_f1_micro": 0.8132532581607222,
|
| 1040 |
+
"eval_loss": 0.12699832022190094,
|
| 1041 |
+
"eval_roc_auc": 0.8770731465947217,
|
| 1042 |
+
"eval_runtime": 538.5901,
|
| 1043 |
+
"eval_samples_per_second": 5.401,
|
| 1044 |
+
"eval_steps_per_second": 0.169,
|
| 1045 |
+
"learning_rate": 0.0001,
|
| 1046 |
+
"step": 16926
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 62.27106227106227,
|
| 1050 |
+
"grad_norm": 0.27737802267074585,
|
| 1051 |
+
"learning_rate": 0.0001,
|
| 1052 |
+
"loss": 0.1299,
|
| 1053 |
+
"step": 17000
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 63.0,
|
| 1057 |
+
"eval_accuracy": 0.29700928154004813,
|
| 1058 |
+
"eval_f1_macro": 0.7059955812986827,
|
| 1059 |
+
"eval_f1_micro": 0.8111204013377926,
|
| 1060 |
+
"eval_loss": 0.12574061751365662,
|
| 1061 |
+
"eval_roc_auc": 0.8692192495020064,
|
| 1062 |
+
"eval_runtime": 534.6194,
|
| 1063 |
+
"eval_samples_per_second": 5.441,
|
| 1064 |
+
"eval_steps_per_second": 0.17,
|
| 1065 |
+
"learning_rate": 0.0001,
|
| 1066 |
+
"step": 17199
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 64.0,
|
| 1070 |
+
"eval_accuracy": 0.2939154348573393,
|
| 1071 |
+
"eval_f1_macro": 0.7090348173219025,
|
| 1072 |
+
"eval_f1_micro": 0.8139229062217472,
|
| 1073 |
+
"eval_loss": 0.1252606362104416,
|
| 1074 |
+
"eval_roc_auc": 0.8757248015353863,
|
| 1075 |
+
"eval_runtime": 540.564,
|
| 1076 |
+
"eval_samples_per_second": 5.381,
|
| 1077 |
+
"eval_steps_per_second": 0.168,
|
| 1078 |
+
"learning_rate": 0.0001,
|
| 1079 |
+
"step": 17472
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 64.1025641025641,
|
| 1083 |
+
"grad_norm": 0.3310890793800354,
|
| 1084 |
+
"learning_rate": 0.0001,
|
| 1085 |
+
"loss": 0.1284,
|
| 1086 |
+
"step": 17500
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 65.0,
|
| 1090 |
+
"eval_accuracy": 0.2963217600550017,
|
| 1091 |
+
"eval_f1_macro": 0.70468515245698,
|
| 1092 |
+
"eval_f1_micro": 0.8126994653292992,
|
| 1093 |
+
"eval_loss": 0.12513236701488495,
|
| 1094 |
+
"eval_roc_auc": 0.8725629317210437,
|
| 1095 |
+
"eval_runtime": 546.3743,
|
| 1096 |
+
"eval_samples_per_second": 5.324,
|
| 1097 |
+
"eval_steps_per_second": 0.167,
|
| 1098 |
+
"learning_rate": 0.0001,
|
| 1099 |
+
"step": 17745
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 65.93406593406593,
|
| 1103 |
+
"grad_norm": 0.2919701635837555,
|
| 1104 |
+
"learning_rate": 0.0001,
|
| 1105 |
+
"loss": 0.1283,
|
| 1106 |
+
"step": 18000
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 66.0,
|
| 1110 |
+
"eval_accuracy": 0.2949467170849089,
|
| 1111 |
+
"eval_f1_macro": 0.7118614262032313,
|
| 1112 |
+
"eval_f1_micro": 0.8140287622403409,
|
| 1113 |
+
"eval_loss": 0.125584214925766,
|
| 1114 |
+
"eval_roc_auc": 0.8766619971126628,
|
| 1115 |
+
"eval_runtime": 538.5562,
|
| 1116 |
+
"eval_samples_per_second": 5.401,
|
| 1117 |
+
"eval_steps_per_second": 0.169,
|
| 1118 |
+
"learning_rate": 0.0001,
|
| 1119 |
+
"step": 18018
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 67.0,
|
| 1123 |
+
"eval_accuracy": 0.2939154348573393,
|
| 1124 |
+
"eval_f1_macro": 0.7062384835160513,
|
| 1125 |
+
"eval_f1_micro": 0.8111171298804116,
|
| 1126 |
+
"eval_loss": 0.12539814412593842,
|
| 1127 |
+
"eval_roc_auc": 0.8701631159427216,
|
| 1128 |
+
"eval_runtime": 547.4711,
|
| 1129 |
+
"eval_samples_per_second": 5.314,
|
| 1130 |
+
"eval_steps_per_second": 0.166,
|
| 1131 |
+
"learning_rate": 0.0001,
|
| 1132 |
+
"step": 18291
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 67.76556776556777,
|
| 1136 |
+
"grad_norm": 0.3583308756351471,
|
| 1137 |
+
"learning_rate": 0.0001,
|
| 1138 |
+
"loss": 0.1281,
|
| 1139 |
+
"step": 18500
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 68.0,
|
| 1143 |
+
"eval_accuracy": 0.29597799931247853,
|
| 1144 |
+
"eval_f1_macro": 0.7112630392285837,
|
| 1145 |
+
"eval_f1_micro": 0.8151627792982313,
|
| 1146 |
+
"eval_loss": 0.12564098834991455,
|
| 1147 |
+
"eval_roc_auc": 0.8781836192989092,
|
| 1148 |
+
"eval_runtime": 544.474,
|
| 1149 |
+
"eval_samples_per_second": 5.343,
|
| 1150 |
+
"eval_steps_per_second": 0.167,
|
| 1151 |
+
"learning_rate": 0.0001,
|
| 1152 |
+
"step": 18564
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 69.0,
|
| 1156 |
+
"eval_accuracy": 0.29941560673771056,
|
| 1157 |
+
"eval_f1_macro": 0.7140636849121678,
|
| 1158 |
+
"eval_f1_micro": 0.8142985980159057,
|
| 1159 |
+
"eval_loss": 0.12584172189235687,
|
| 1160 |
+
"eval_roc_auc": 0.8766551190920747,
|
| 1161 |
+
"eval_runtime": 544.6719,
|
| 1162 |
+
"eval_samples_per_second": 5.341,
|
| 1163 |
+
"eval_steps_per_second": 0.167,
|
| 1164 |
+
"learning_rate": 0.0001,
|
| 1165 |
+
"step": 18837
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 69.59706959706959,
|
| 1169 |
+
"grad_norm": 0.30125734210014343,
|
| 1170 |
+
"learning_rate": 0.0001,
|
| 1171 |
+
"loss": 0.1274,
|
| 1172 |
+
"step": 19000
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 70.0,
|
| 1176 |
+
"eval_accuracy": 0.29838432451014096,
|
| 1177 |
+
"eval_f1_macro": 0.718410533722422,
|
| 1178 |
+
"eval_f1_micro": 0.8161761696205642,
|
| 1179 |
+
"eval_loss": 0.12478043138980865,
|
| 1180 |
+
"eval_roc_auc": 0.8781431798039784,
|
| 1181 |
+
"eval_runtime": 533.1364,
|
| 1182 |
+
"eval_samples_per_second": 5.456,
|
| 1183 |
+
"eval_steps_per_second": 0.171,
|
| 1184 |
+
"learning_rate": 0.0001,
|
| 1185 |
+
"step": 19110
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 71.0,
|
| 1189 |
+
"eval_accuracy": 0.2963217600550017,
|
| 1190 |
+
"eval_f1_macro": 0.7097253169772126,
|
| 1191 |
+
"eval_f1_micro": 0.8134349886668041,
|
| 1192 |
+
"eval_loss": 0.12550216913223267,
|
| 1193 |
+
"eval_roc_auc": 0.8746253918294316,
|
| 1194 |
+
"eval_runtime": 531.413,
|
| 1195 |
+
"eval_samples_per_second": 5.474,
|
| 1196 |
+
"eval_steps_per_second": 0.171,
|
| 1197 |
+
"learning_rate": 0.0001,
|
| 1198 |
+
"step": 19383
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 71.42857142857143,
|
| 1202 |
+
"grad_norm": 0.2948232591152191,
|
| 1203 |
+
"learning_rate": 0.0001,
|
| 1204 |
+
"loss": 0.1274,
|
| 1205 |
+
"step": 19500
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 72.0,
|
| 1209 |
+
"eval_accuracy": 0.2980405637676177,
|
| 1210 |
+
"eval_f1_macro": 0.7066535688303756,
|
| 1211 |
+
"eval_f1_micro": 0.8132986082851795,
|
| 1212 |
+
"eval_loss": 0.12681567668914795,
|
| 1213 |
+
"eval_roc_auc": 0.8782642911316215,
|
| 1214 |
+
"eval_runtime": 530.4686,
|
| 1215 |
+
"eval_samples_per_second": 5.484,
|
| 1216 |
+
"eval_steps_per_second": 0.172,
|
| 1217 |
+
"learning_rate": 0.0001,
|
| 1218 |
+
"step": 19656
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 73.0,
|
| 1222 |
+
"eval_accuracy": 0.30044688896528016,
|
| 1223 |
+
"eval_f1_macro": 0.720052479680047,
|
| 1224 |
+
"eval_f1_micro": 0.8154191311441974,
|
| 1225 |
+
"eval_loss": 0.12942491471767426,
|
| 1226 |
+
"eval_roc_auc": 0.8787365560050311,
|
| 1227 |
+
"eval_runtime": 542.0988,
|
| 1228 |
+
"eval_samples_per_second": 5.366,
|
| 1229 |
+
"eval_steps_per_second": 0.168,
|
| 1230 |
+
"learning_rate": 0.0001,
|
| 1231 |
+
"step": 19929
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 73.26007326007326,
|
| 1235 |
+
"grad_norm": 0.36606550216674805,
|
| 1236 |
+
"learning_rate": 0.0001,
|
| 1237 |
+
"loss": 0.1272,
|
| 1238 |
+
"step": 20000
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 74.0,
|
| 1242 |
+
"eval_accuracy": 0.29597799931247853,
|
| 1243 |
+
"eval_f1_macro": 0.7156506533830808,
|
| 1244 |
+
"eval_f1_micro": 0.8141104799538981,
|
| 1245 |
+
"eval_loss": 0.1252431720495224,
|
| 1246 |
+
"eval_roc_auc": 0.8753470063705208,
|
| 1247 |
+
"eval_runtime": 537.4536,
|
| 1248 |
+
"eval_samples_per_second": 5.413,
|
| 1249 |
+
"eval_steps_per_second": 0.169,
|
| 1250 |
+
"learning_rate": 0.0001,
|
| 1251 |
+
"step": 20202
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 75.0,
|
| 1255 |
+
"eval_accuracy": 0.29975936748023374,
|
| 1256 |
+
"eval_f1_macro": 0.7162358218470065,
|
| 1257 |
+
"eval_f1_micro": 0.8164923076923079,
|
| 1258 |
+
"eval_loss": 0.12525109946727753,
|
| 1259 |
+
"eval_roc_auc": 0.877641538433871,
|
| 1260 |
+
"eval_runtime": 537.3121,
|
| 1261 |
+
"eval_samples_per_second": 5.414,
|
| 1262 |
+
"eval_steps_per_second": 0.169,
|
| 1263 |
+
"learning_rate": 0.0001,
|
| 1264 |
+
"step": 20475
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 75.0915750915751,
|
| 1268 |
+
"grad_norm": 0.3238458037376404,
|
| 1269 |
+
"learning_rate": 0.0001,
|
| 1270 |
+
"loss": 0.1265,
|
| 1271 |
+
"step": 20500
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 76.0,
|
| 1275 |
+
"eval_accuracy": 0.2980405637676177,
|
| 1276 |
+
"eval_f1_macro": 0.7050812600051921,
|
| 1277 |
+
"eval_f1_micro": 0.8126263668248401,
|
| 1278 |
+
"eval_loss": 0.125084787607193,
|
| 1279 |
+
"eval_roc_auc": 0.8738193347922841,
|
| 1280 |
+
"eval_runtime": 528.5209,
|
| 1281 |
+
"eval_samples_per_second": 5.504,
|
| 1282 |
+
"eval_steps_per_second": 0.172,
|
| 1283 |
+
"learning_rate": 0.0001,
|
| 1284 |
+
"step": 20748
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 76.92307692307692,
|
| 1288 |
+
"grad_norm": 0.3312053382396698,
|
| 1289 |
+
"learning_rate": 1e-05,
|
| 1290 |
+
"loss": 0.1253,
|
| 1291 |
+
"step": 21000
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 77.0,
|
| 1295 |
+
"eval_accuracy": 0.301821931935373,
|
| 1296 |
+
"eval_f1_macro": 0.7198797063603358,
|
| 1297 |
+
"eval_f1_micro": 0.8169637369391518,
|
| 1298 |
+
"eval_loss": 0.12381099909543991,
|
| 1299 |
+
"eval_roc_auc": 0.8782657830663234,
|
| 1300 |
+
"eval_runtime": 534.6595,
|
| 1301 |
+
"eval_samples_per_second": 5.441,
|
| 1302 |
+
"eval_steps_per_second": 0.17,
|
| 1303 |
+
"learning_rate": 1e-05,
|
| 1304 |
+
"step": 21021
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 78.0,
|
| 1308 |
+
"eval_accuracy": 0.29975936748023374,
|
| 1309 |
+
"eval_f1_macro": 0.7125350103590883,
|
| 1310 |
+
"eval_f1_micro": 0.8142230317079229,
|
| 1311 |
+
"eval_loss": 0.12425024807453156,
|
| 1312 |
+
"eval_roc_auc": 0.8737716029624661,
|
| 1313 |
+
"eval_runtime": 531.5279,
|
| 1314 |
+
"eval_samples_per_second": 5.473,
|
| 1315 |
+
"eval_steps_per_second": 0.171,
|
| 1316 |
+
"learning_rate": 1e-05,
|
| 1317 |
+
"step": 21294
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 78.75457875457876,
|
| 1321 |
+
"grad_norm": 0.3140043020248413,
|
| 1322 |
+
"learning_rate": 1e-05,
|
| 1323 |
+
"loss": 0.1231,
|
| 1324 |
+
"step": 21500
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 79.0,
|
| 1328 |
+
"eval_accuracy": 0.3028532141629426,
|
| 1329 |
+
"eval_f1_macro": 0.721102526527663,
|
| 1330 |
+
"eval_f1_micro": 0.8181743958197256,
|
| 1331 |
+
"eval_loss": 0.12345358729362488,
|
| 1332 |
+
"eval_roc_auc": 0.8800411470550439,
|
| 1333 |
+
"eval_runtime": 529.7514,
|
| 1334 |
+
"eval_samples_per_second": 5.491,
|
| 1335 |
+
"eval_steps_per_second": 0.172,
|
| 1336 |
+
"learning_rate": 1e-05,
|
| 1337 |
+
"step": 21567
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 80.0,
|
| 1341 |
+
"eval_accuracy": 0.30216569267789617,
|
| 1342 |
+
"eval_f1_macro": 0.7272188807325698,
|
| 1343 |
+
"eval_f1_micro": 0.8199023445381542,
|
| 1344 |
+
"eval_loss": 0.12395191192626953,
|
| 1345 |
+
"eval_roc_auc": 0.8844715187246072,
|
| 1346 |
+
"eval_runtime": 537.5403,
|
| 1347 |
+
"eval_samples_per_second": 5.412,
|
| 1348 |
+
"eval_steps_per_second": 0.169,
|
| 1349 |
+
"learning_rate": 1e-05,
|
| 1350 |
+
"step": 21840
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 80.58608058608058,
|
| 1354 |
+
"grad_norm": 0.2950115203857422,
|
| 1355 |
+
"learning_rate": 1e-05,
|
| 1356 |
+
"loss": 0.1232,
|
| 1357 |
+
"step": 22000
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 81.0,
|
| 1361 |
+
"eval_accuracy": 0.3011344104503266,
|
| 1362 |
+
"eval_f1_macro": 0.7213714212722541,
|
| 1363 |
+
"eval_f1_micro": 0.816331575477917,
|
| 1364 |
+
"eval_loss": 0.12421117722988129,
|
| 1365 |
+
"eval_roc_auc": 0.876307363730011,
|
| 1366 |
+
"eval_runtime": 531.2157,
|
| 1367 |
+
"eval_samples_per_second": 5.476,
|
| 1368 |
+
"eval_steps_per_second": 0.171,
|
| 1369 |
+
"learning_rate": 1e-05,
|
| 1370 |
+
"step": 22113
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 82.0,
|
| 1374 |
+
"eval_accuracy": 0.3042282571330354,
|
| 1375 |
+
"eval_f1_macro": 0.7220641511602056,
|
| 1376 |
+
"eval_f1_micro": 0.8184078588024294,
|
| 1377 |
+
"eval_loss": 0.1238287091255188,
|
| 1378 |
+
"eval_roc_auc": 0.8806201457434719,
|
| 1379 |
+
"eval_runtime": 539.7944,
|
| 1380 |
+
"eval_samples_per_second": 5.389,
|
| 1381 |
+
"eval_steps_per_second": 0.169,
|
| 1382 |
+
"learning_rate": 1e-05,
|
| 1383 |
+
"step": 22386
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 82.41758241758242,
|
| 1387 |
+
"grad_norm": 0.3993464708328247,
|
| 1388 |
+
"learning_rate": 1e-05,
|
| 1389 |
+
"loss": 0.1223,
|
| 1390 |
+
"step": 22500
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 83.0,
|
| 1394 |
+
"eval_accuracy": 0.301821931935373,
|
| 1395 |
+
"eval_f1_macro": 0.7213144754619842,
|
| 1396 |
+
"eval_f1_micro": 0.8180256808702052,
|
| 1397 |
+
"eval_loss": 0.1235181912779808,
|
| 1398 |
+
"eval_roc_auc": 0.8794490137968879,
|
| 1399 |
+
"eval_runtime": 534.2182,
|
| 1400 |
+
"eval_samples_per_second": 5.445,
|
| 1401 |
+
"eval_steps_per_second": 0.17,
|
| 1402 |
+
"learning_rate": 1e-05,
|
| 1403 |
+
"step": 22659
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 84.0,
|
| 1407 |
+
"eval_accuracy": 0.3025094534204194,
|
| 1408 |
+
"eval_f1_macro": 0.7232957874603048,
|
| 1409 |
+
"eval_f1_micro": 0.8176539851394697,
|
| 1410 |
+
"eval_loss": 0.12367285788059235,
|
| 1411 |
+
"eval_roc_auc": 0.8813632229749234,
|
| 1412 |
+
"eval_runtime": 527.4506,
|
| 1413 |
+
"eval_samples_per_second": 5.515,
|
| 1414 |
+
"eval_steps_per_second": 0.173,
|
| 1415 |
+
"learning_rate": 1e-05,
|
| 1416 |
+
"step": 22932
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 84.24908424908425,
|
| 1420 |
+
"grad_norm": 0.2845182716846466,
|
| 1421 |
+
"learning_rate": 1e-05,
|
| 1422 |
+
"loss": 0.1216,
|
| 1423 |
+
"step": 23000
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 85.0,
|
| 1427 |
+
"eval_accuracy": 0.3045720178755586,
|
| 1428 |
+
"eval_f1_macro": 0.7142044524891776,
|
| 1429 |
+
"eval_f1_micro": 0.8164039937288555,
|
| 1430 |
+
"eval_loss": 0.12321745604276657,
|
| 1431 |
+
"eval_roc_auc": 0.8759362394285842,
|
| 1432 |
+
"eval_runtime": 530.9606,
|
| 1433 |
+
"eval_samples_per_second": 5.479,
|
| 1434 |
+
"eval_steps_per_second": 0.171,
|
| 1435 |
+
"learning_rate": 1e-05,
|
| 1436 |
+
"step": 23205
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 86.0,
|
| 1440 |
+
"eval_accuracy": 0.303540735647989,
|
| 1441 |
+
"eval_f1_macro": 0.724290897047381,
|
| 1442 |
+
"eval_f1_micro": 0.8165334212478365,
|
| 1443 |
+
"eval_loss": 0.1237027570605278,
|
| 1444 |
+
"eval_roc_auc": 0.8763461459677461,
|
| 1445 |
+
"eval_runtime": 529.6733,
|
| 1446 |
+
"eval_samples_per_second": 5.492,
|
| 1447 |
+
"eval_steps_per_second": 0.172,
|
| 1448 |
+
"learning_rate": 1e-05,
|
| 1449 |
+
"step": 23478
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 86.08058608058609,
|
| 1453 |
+
"grad_norm": 0.3410782516002655,
|
| 1454 |
+
"learning_rate": 1e-05,
|
| 1455 |
+
"loss": 0.1217,
|
| 1456 |
+
"step": 23500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 87.0,
|
| 1460 |
+
"eval_accuracy": 0.30216569267789617,
|
| 1461 |
+
"eval_f1_macro": 0.7200937974757147,
|
| 1462 |
+
"eval_f1_micro": 0.8180700172173485,
|
| 1463 |
+
"eval_loss": 0.12352145463228226,
|
| 1464 |
+
"eval_roc_auc": 0.8787459617460684,
|
| 1465 |
+
"eval_runtime": 538.921,
|
| 1466 |
+
"eval_samples_per_second": 5.398,
|
| 1467 |
+
"eval_steps_per_second": 0.169,
|
| 1468 |
+
"learning_rate": 1e-05,
|
| 1469 |
+
"step": 23751
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 87.91208791208791,
|
| 1473 |
+
"grad_norm": 0.33241167664527893,
|
| 1474 |
+
"learning_rate": 1e-05,
|
| 1475 |
+
"loss": 0.1206,
|
| 1476 |
+
"step": 24000
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 88.0,
|
| 1480 |
+
"eval_accuracy": 0.303540735647989,
|
| 1481 |
+
"eval_f1_macro": 0.7208874018784168,
|
| 1482 |
+
"eval_f1_micro": 0.8187826933214387,
|
| 1483 |
+
"eval_loss": 0.12393338233232498,
|
| 1484 |
+
"eval_roc_auc": 0.8823836181053615,
|
| 1485 |
+
"eval_runtime": 527.0575,
|
| 1486 |
+
"eval_samples_per_second": 5.519,
|
| 1487 |
+
"eval_steps_per_second": 0.173,
|
| 1488 |
+
"learning_rate": 1e-05,
|
| 1489 |
+
"step": 24024
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 89.0,
|
| 1493 |
+
"eval_accuracy": 0.3028532141629426,
|
| 1494 |
+
"eval_f1_macro": 0.717195490689832,
|
| 1495 |
+
"eval_f1_micro": 0.8175708900180297,
|
| 1496 |
+
"eval_loss": 0.12368057668209076,
|
| 1497 |
+
"eval_roc_auc": 0.8785904184808291,
|
| 1498 |
+
"eval_runtime": 530.0815,
|
| 1499 |
+
"eval_samples_per_second": 5.488,
|
| 1500 |
+
"eval_steps_per_second": 0.172,
|
| 1501 |
+
"learning_rate": 1e-05,
|
| 1502 |
+
"step": 24297
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 89.74358974358974,
|
| 1506 |
+
"grad_norm": 0.33269670605659485,
|
| 1507 |
+
"learning_rate": 1e-05,
|
| 1508 |
+
"loss": 0.1215,
|
| 1509 |
+
"step": 24500
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 90.0,
|
| 1513 |
+
"eval_accuracy": 0.30594706084565143,
|
| 1514 |
+
"eval_f1_macro": 0.7202869112854988,
|
| 1515 |
+
"eval_f1_micro": 0.8169988469774335,
|
| 1516 |
+
"eval_loss": 0.12319833785295486,
|
| 1517 |
+
"eval_roc_auc": 0.8768206895697999,
|
| 1518 |
+
"eval_runtime": 527.39,
|
| 1519 |
+
"eval_samples_per_second": 5.516,
|
| 1520 |
+
"eval_steps_per_second": 0.173,
|
| 1521 |
+
"learning_rate": 1e-05,
|
| 1522 |
+
"step": 24570
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 91.0,
|
| 1526 |
+
"eval_accuracy": 0.303540735647989,
|
| 1527 |
+
"eval_f1_macro": 0.7243798285402773,
|
| 1528 |
+
"eval_f1_micro": 0.8199162022535897,
|
| 1529 |
+
"eval_loss": 0.12350083887577057,
|
| 1530 |
+
"eval_roc_auc": 0.88206963137479,
|
| 1531 |
+
"eval_runtime": 529.1998,
|
| 1532 |
+
"eval_samples_per_second": 5.497,
|
| 1533 |
+
"eval_steps_per_second": 0.172,
|
| 1534 |
+
"learning_rate": 1e-05,
|
| 1535 |
+
"step": 24843
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 91.57509157509158,
|
| 1539 |
+
"grad_norm": 0.2958417534828186,
|
| 1540 |
+
"learning_rate": 1e-05,
|
| 1541 |
+
"loss": 0.1216,
|
| 1542 |
+
"step": 25000
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 92.0,
|
| 1546 |
+
"eval_accuracy": 0.30560330010312825,
|
| 1547 |
+
"eval_f1_macro": 0.7264331892084873,
|
| 1548 |
+
"eval_f1_micro": 0.820052770448549,
|
| 1549 |
+
"eval_loss": 0.12410824745893478,
|
| 1550 |
+
"eval_roc_auc": 0.8827789399747492,
|
| 1551 |
+
"eval_runtime": 533.4656,
|
| 1552 |
+
"eval_samples_per_second": 5.453,
|
| 1553 |
+
"eval_steps_per_second": 0.171,
|
| 1554 |
+
"learning_rate": 1e-05,
|
| 1555 |
+
"step": 25116
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 93.0,
|
| 1559 |
+
"eval_accuracy": 0.3028532141629426,
|
| 1560 |
+
"eval_f1_macro": 0.7265708763385225,
|
| 1561 |
+
"eval_f1_micro": 0.8191851972082453,
|
| 1562 |
+
"eval_loss": 0.12333476543426514,
|
| 1563 |
+
"eval_roc_auc": 0.8824025949099301,
|
| 1564 |
+
"eval_runtime": 530.3618,
|
| 1565 |
+
"eval_samples_per_second": 5.485,
|
| 1566 |
+
"eval_steps_per_second": 0.172,
|
| 1567 |
+
"learning_rate": 1e-05,
|
| 1568 |
+
"step": 25389
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 93.4065934065934,
|
| 1572 |
+
"grad_norm": 0.2711620032787323,
|
| 1573 |
+
"learning_rate": 1e-05,
|
| 1574 |
+
"loss": 0.1203,
|
| 1575 |
+
"step": 25500
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 94.0,
|
| 1579 |
+
"eval_accuracy": 0.29769680302509455,
|
| 1580 |
+
"eval_f1_macro": 0.7139339929197925,
|
| 1581 |
+
"eval_f1_micro": 0.8147507922788823,
|
| 1582 |
+
"eval_loss": 0.12396726757287979,
|
| 1583 |
+
"eval_roc_auc": 0.8757366931622173,
|
| 1584 |
+
"eval_runtime": 526.1521,
|
| 1585 |
+
"eval_samples_per_second": 5.529,
|
| 1586 |
+
"eval_steps_per_second": 0.173,
|
| 1587 |
+
"learning_rate": 1e-05,
|
| 1588 |
+
"step": 25662
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 95.0,
|
| 1592 |
+
"eval_accuracy": 0.2980405637676177,
|
| 1593 |
+
"eval_f1_macro": 0.7272882506355968,
|
| 1594 |
+
"eval_f1_micro": 0.8202541859995964,
|
| 1595 |
+
"eval_loss": 0.12366786599159241,
|
| 1596 |
+
"eval_roc_auc": 0.8847183825064932,
|
| 1597 |
+
"eval_runtime": 532.6436,
|
| 1598 |
+
"eval_samples_per_second": 5.461,
|
| 1599 |
+
"eval_steps_per_second": 0.171,
|
| 1600 |
+
"learning_rate": 1e-05,
|
| 1601 |
+
"step": 25935
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 95.23809523809524,
|
| 1605 |
+
"grad_norm": 0.34573543071746826,
|
| 1606 |
+
"learning_rate": 1e-05,
|
| 1607 |
+
"loss": 0.1215,
|
| 1608 |
+
"step": 26000
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 96.0,
|
| 1612 |
+
"eval_accuracy": 0.30491577861808183,
|
| 1613 |
+
"eval_f1_macro": 0.7202747551284554,
|
| 1614 |
+
"eval_f1_micro": 0.8192505510653931,
|
| 1615 |
+
"eval_loss": 0.12370884418487549,
|
| 1616 |
+
"eval_roc_auc": 0.8806645630588985,
|
| 1617 |
+
"eval_runtime": 529.0949,
|
| 1618 |
+
"eval_samples_per_second": 5.498,
|
| 1619 |
+
"eval_steps_per_second": 0.172,
|
| 1620 |
+
"learning_rate": 1e-05,
|
| 1621 |
+
"step": 26208
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 97.0,
|
| 1625 |
+
"eval_accuracy": 0.3007906497078034,
|
| 1626 |
+
"eval_f1_macro": 0.7212605213289621,
|
| 1627 |
+
"eval_f1_micro": 0.8185695138296577,
|
| 1628 |
+
"eval_loss": 0.12347108125686646,
|
| 1629 |
+
"eval_roc_auc": 0.8817785574346272,
|
| 1630 |
+
"eval_runtime": 536.573,
|
| 1631 |
+
"eval_samples_per_second": 5.421,
|
| 1632 |
+
"eval_steps_per_second": 0.17,
|
| 1633 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1634 |
+
"step": 26481
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 97.06959706959707,
|
| 1638 |
+
"grad_norm": 0.42561525106430054,
|
| 1639 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1640 |
+
"loss": 0.1211,
|
| 1641 |
+
"step": 26500
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 98.0,
|
| 1645 |
+
"eval_accuracy": 0.3038844963905122,
|
| 1646 |
+
"eval_f1_macro": 0.7173148781861652,
|
| 1647 |
+
"eval_f1_micro": 0.8167741405511973,
|
| 1648 |
+
"eval_loss": 0.12309076637029648,
|
| 1649 |
+
"eval_roc_auc": 0.8774588974005779,
|
| 1650 |
+
"eval_runtime": 542.2112,
|
| 1651 |
+
"eval_samples_per_second": 5.365,
|
| 1652 |
+
"eval_steps_per_second": 0.168,
|
| 1653 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1654 |
+
"step": 26754
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 98.9010989010989,
|
| 1658 |
+
"grad_norm": 0.3780280351638794,
|
| 1659 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1660 |
+
"loss": 0.1202,
|
| 1661 |
+
"step": 27000
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 99.0,
|
| 1665 |
+
"eval_accuracy": 0.30216569267789617,
|
| 1666 |
+
"eval_f1_macro": 0.7265958923493796,
|
| 1667 |
+
"eval_f1_micro": 0.8195458231954581,
|
| 1668 |
+
"eval_loss": 0.12337860465049744,
|
| 1669 |
+
"eval_roc_auc": 0.8827991597222147,
|
| 1670 |
+
"eval_runtime": 542.1069,
|
| 1671 |
+
"eval_samples_per_second": 5.366,
|
| 1672 |
+
"eval_steps_per_second": 0.168,
|
| 1673 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1674 |
+
"step": 27027
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 100.0,
|
| 1678 |
+
"eval_accuracy": 0.3025094534204194,
|
| 1679 |
+
"eval_f1_macro": 0.721153078194628,
|
| 1680 |
+
"eval_f1_micro": 0.8176836250613447,
|
| 1681 |
+
"eval_loss": 0.12339676916599274,
|
| 1682 |
+
"eval_roc_auc": 0.8792346757035974,
|
| 1683 |
+
"eval_runtime": 536.027,
|
| 1684 |
+
"eval_samples_per_second": 5.427,
|
| 1685 |
+
"eval_steps_per_second": 0.17,
|
| 1686 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1687 |
+
"step": 27300
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 100.73260073260073,
|
| 1691 |
+
"grad_norm": 0.2755577862262726,
|
| 1692 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1693 |
+
"loss": 0.1202,
|
| 1694 |
+
"step": 27500
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 101.0,
|
| 1698 |
+
"eval_accuracy": 0.3007906497078034,
|
| 1699 |
+
"eval_f1_macro": 0.727514273407078,
|
| 1700 |
+
"eval_f1_micro": 0.8195522327414572,
|
| 1701 |
+
"eval_loss": 0.12300820648670197,
|
| 1702 |
+
"eval_roc_auc": 0.8822068193069091,
|
| 1703 |
+
"eval_runtime": 540.6402,
|
| 1704 |
+
"eval_samples_per_second": 5.381,
|
| 1705 |
+
"eval_steps_per_second": 0.168,
|
| 1706 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1707 |
+
"step": 27573
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 102.0,
|
| 1711 |
+
"eval_accuracy": 0.30147817119284975,
|
| 1712 |
+
"eval_f1_macro": 0.7137375786593034,
|
| 1713 |
+
"eval_f1_micro": 0.8154413898909484,
|
| 1714 |
+
"eval_loss": 0.12340469658374786,
|
| 1715 |
+
"eval_roc_auc": 0.8739457087179753,
|
| 1716 |
+
"eval_runtime": 543.6517,
|
| 1717 |
+
"eval_samples_per_second": 5.351,
|
| 1718 |
+
"eval_steps_per_second": 0.167,
|
| 1719 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1720 |
+
"step": 27846
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 102.56410256410257,
|
| 1724 |
+
"grad_norm": 0.41754451394081116,
|
| 1725 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1726 |
+
"loss": 0.1212,
|
| 1727 |
+
"step": 28000
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 103.0,
|
| 1731 |
+
"eval_accuracy": 0.29941560673771056,
|
| 1732 |
+
"eval_f1_macro": 0.7145427089697738,
|
| 1733 |
+
"eval_f1_micro": 0.8163972286374134,
|
| 1734 |
+
"eval_loss": 0.12386961281299591,
|
| 1735 |
+
"eval_roc_auc": 0.8760534147703873,
|
| 1736 |
+
"eval_runtime": 542.2436,
|
| 1737 |
+
"eval_samples_per_second": 5.365,
|
| 1738 |
+
"eval_steps_per_second": 0.168,
|
| 1739 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1740 |
+
"step": 28119
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 104.0,
|
| 1744 |
+
"eval_accuracy": 0.30147817119284975,
|
| 1745 |
+
"eval_f1_macro": 0.7188862048208717,
|
| 1746 |
+
"eval_f1_micro": 0.8170896715732502,
|
| 1747 |
+
"eval_loss": 0.1235337182879448,
|
| 1748 |
+
"eval_roc_auc": 0.8781419368610816,
|
| 1749 |
+
"eval_runtime": 543.5285,
|
| 1750 |
+
"eval_samples_per_second": 5.352,
|
| 1751 |
+
"eval_steps_per_second": 0.167,
|
| 1752 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1753 |
+
"step": 28392
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 104.3956043956044,
|
| 1757 |
+
"grad_norm": 0.33403730392456055,
|
| 1758 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1759 |
+
"loss": 0.1205,
|
| 1760 |
+
"step": 28500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 105.0,
|
| 1764 |
+
"eval_accuracy": 0.301821931935373,
|
| 1765 |
+
"eval_f1_macro": 0.7225140266508299,
|
| 1766 |
+
"eval_f1_micro": 0.8191040415516962,
|
| 1767 |
+
"eval_loss": 0.12342803180217743,
|
| 1768 |
+
"eval_roc_auc": 0.8823571418087561,
|
| 1769 |
+
"eval_runtime": 543.3377,
|
| 1770 |
+
"eval_samples_per_second": 5.354,
|
| 1771 |
+
"eval_steps_per_second": 0.167,
|
| 1772 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1773 |
+
"step": 28665
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 106.0,
|
| 1777 |
+
"eval_accuracy": 0.30216569267789617,
|
| 1778 |
+
"eval_f1_macro": 0.7182648261775325,
|
| 1779 |
+
"eval_f1_micro": 0.8179164977705716,
|
| 1780 |
+
"eval_loss": 0.12395947426557541,
|
| 1781 |
+
"eval_roc_auc": 0.8820072730405425,
|
| 1782 |
+
"eval_runtime": 543.5076,
|
| 1783 |
+
"eval_samples_per_second": 5.352,
|
| 1784 |
+
"eval_steps_per_second": 0.167,
|
| 1785 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1786 |
+
"step": 28938
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 106.22710622710623,
|
| 1790 |
+
"grad_norm": 0.2938772439956665,
|
| 1791 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1792 |
+
"loss": 0.1209,
|
| 1793 |
+
"step": 29000
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 107.0,
|
| 1797 |
+
"eval_accuracy": 0.30147817119284975,
|
| 1798 |
+
"eval_f1_macro": 0.7269935921863244,
|
| 1799 |
+
"eval_f1_micro": 0.8188311688311688,
|
| 1800 |
+
"eval_loss": 0.1234135553240776,
|
| 1801 |
+
"eval_roc_auc": 0.882155731128044,
|
| 1802 |
+
"eval_runtime": 539.5012,
|
| 1803 |
+
"eval_samples_per_second": 5.392,
|
| 1804 |
+
"eval_steps_per_second": 0.169,
|
| 1805 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1806 |
+
"step": 29211
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 108.0,
|
| 1810 |
+
"eval_accuracy": 0.29975936748023374,
|
| 1811 |
+
"eval_f1_macro": 0.722601561126108,
|
| 1812 |
+
"eval_f1_micro": 0.8162340337865678,
|
| 1813 |
+
"eval_loss": 0.12344102561473846,
|
| 1814 |
+
"eval_roc_auc": 0.8762293849402418,
|
| 1815 |
+
"eval_runtime": 553.8091,
|
| 1816 |
+
"eval_samples_per_second": 5.253,
|
| 1817 |
+
"eval_steps_per_second": 0.164,
|
| 1818 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 1819 |
+
"step": 29484
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"epoch": 108.05860805860806,
|
| 1823 |
+
"grad_norm": 0.3646715581417084,
|
| 1824 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 1825 |
+
"loss": 0.1201,
|
| 1826 |
+
"step": 29500
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 109.0,
|
| 1830 |
+
"eval_accuracy": 0.303540735647989,
|
| 1831 |
+
"eval_f1_macro": 0.7257459935415149,
|
| 1832 |
+
"eval_f1_micro": 0.8215820979470492,
|
| 1833 |
+
"eval_loss": 0.12361280620098114,
|
| 1834 |
+
"eval_roc_auc": 0.886760415710266,
|
| 1835 |
+
"eval_runtime": 547.1195,
|
| 1836 |
+
"eval_samples_per_second": 5.317,
|
| 1837 |
+
"eval_steps_per_second": 0.166,
|
| 1838 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 1839 |
+
"step": 29757
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 109.89010989010988,
|
| 1843 |
+
"grad_norm": 0.3803616464138031,
|
| 1844 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 1845 |
+
"loss": 0.1205,
|
| 1846 |
+
"step": 30000
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 110.0,
|
| 1850 |
+
"eval_accuracy": 0.3031969749054658,
|
| 1851 |
+
"eval_f1_macro": 0.7232171570839454,
|
| 1852 |
+
"eval_f1_micro": 0.8204050284975141,
|
| 1853 |
+
"eval_loss": 0.12370219826698303,
|
| 1854 |
+
"eval_roc_auc": 0.8842430102758414,
|
| 1855 |
+
"eval_runtime": 543.9694,
|
| 1856 |
+
"eval_samples_per_second": 5.348,
|
| 1857 |
+
"eval_steps_per_second": 0.167,
|
| 1858 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 1859 |
+
"step": 30030
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 111.0,
|
| 1863 |
+
"eval_accuracy": 0.305259539360605,
|
| 1864 |
+
"eval_f1_macro": 0.7223581742811798,
|
| 1865 |
+
"eval_f1_micro": 0.8178334500803495,
|
| 1866 |
+
"eval_loss": 0.12323758751153946,
|
| 1867 |
+
"eval_roc_auc": 0.8770994575688328,
|
| 1868 |
+
"eval_runtime": 543.4841,
|
| 1869 |
+
"eval_samples_per_second": 5.353,
|
| 1870 |
+
"eval_steps_per_second": 0.167,
|
| 1871 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 1872 |
+
"step": 30303
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 111.0,
|
| 1876 |
+
"learning_rate": 1.0000000000000002e-07,
|
| 1877 |
+
"step": 30303,
|
| 1878 |
+
"total_flos": 1.4332659431723232e+21,
|
| 1879 |
+
"train_loss": 0.13822040822849538,
|
| 1880 |
+
"train_runtime": 237956.1018,
|
| 1881 |
+
"train_samples_per_second": 5.489,
|
| 1882 |
+
"train_steps_per_second": 0.172
|
| 1883 |
+
}
|
| 1884 |
+
],
|
| 1885 |
+
"logging_steps": 500,
|
| 1886 |
+
"max_steps": 40950,
|
| 1887 |
+
"num_input_tokens_seen": 0,
|
| 1888 |
+
"num_train_epochs": 150,
|
| 1889 |
+
"save_steps": 500,
|
| 1890 |
+
"stateful_callbacks": {
|
| 1891 |
+
"EarlyStoppingCallback": {
|
| 1892 |
+
"args": {
|
| 1893 |
+
"early_stopping_patience": 10,
|
| 1894 |
+
"early_stopping_threshold": 0.0
|
| 1895 |
+
},
|
| 1896 |
+
"attributes": {
|
| 1897 |
+
"early_stopping_patience_counter": 0
|
| 1898 |
+
}
|
| 1899 |
+
},
|
| 1900 |
+
"TrainerControl": {
|
| 1901 |
+
"args": {
|
| 1902 |
+
"should_epoch_stop": false,
|
| 1903 |
+
"should_evaluate": false,
|
| 1904 |
+
"should_log": false,
|
| 1905 |
+
"should_save": true,
|
| 1906 |
+
"should_training_stop": true
|
| 1907 |
+
},
|
| 1908 |
+
"attributes": {}
|
| 1909 |
+
}
|
| 1910 |
+
},
|
| 1911 |
+
"total_flos": 1.4332659431723232e+21,
|
| 1912 |
+
"train_batch_size": 32,
|
| 1913 |
+
"trial_name": null,
|
| 1914 |
+
"trial_params": null
|
| 1915 |
+
}
|