t5-base_sst2_dense

This model is a fine-tuned version of t5-base on the glue dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2156
  • Accuracy: 0.9232

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 32
  • eval_batch_size: 64
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 200
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6905 0.01 10 0.7366 0.5080
0.684 0.02 20 0.7306 0.5069
0.7013 0.03 30 0.7228 0.5080
0.6954 0.04 40 0.7114 0.5046
0.6893 0.05 50 0.7026 0.5034
0.6888 0.06 60 0.6912 0.5023
0.6814 0.07 70 0.6848 0.5034
0.679 0.08 80 0.6745 0.5206
0.6616 0.09 90 0.6685 0.5252
0.6604 0.1 100 0.6580 0.5378
0.6524 0.1 110 0.6378 0.6525
0.6344 0.11 120 0.6128 0.7271
0.5915 0.12 130 0.5672 0.8016
0.562 0.13 140 0.4903 0.8578
0.4653 0.14 150 0.3825 0.8796
0.3632 0.15 160 0.2811 0.8991
0.2754 0.16 170 0.3029 0.8933
0.2298 0.17 180 0.3001 0.8991
0.2819 0.18 190 0.2636 0.9083
0.2532 0.19 200 0.2321 0.9128
0.2512 0.2 210 0.2286 0.9186
0.2149 0.21 220 0.2424 0.9128
0.2466 0.22 230 0.2505 0.9140
0.1853 0.23 240 0.2178 0.9186
0.2279 0.24 250 0.2152 0.9186
0.219 0.25 260 0.2188 0.9197
0.2144 0.26 270 0.2179 0.9209
0.1507 0.27 280 0.2185 0.9186
0.1801 0.28 290 0.2473 0.9243
0.1735 0.29 300 0.2402 0.9128
0.1437 0.29 310 0.2436 0.9255
0.2221 0.3 320 0.2209 0.9163
0.1611 0.31 330 0.2101 0.9232
0.1813 0.32 340 0.2291 0.9174
0.1871 0.33 350 0.2386 0.9174
0.2126 0.34 360 0.2225 0.9197
0.2023 0.35 370 0.2116 0.9232
0.127 0.36 380 0.2155 0.9232
0.2769 0.37 390 0.2149 0.9243
0.1457 0.38 400 0.2166 0.9232
0.2129 0.39 410 0.2271 0.9232
0.1652 0.4 420 0.2308 0.9220
0.1783 0.41 430 0.2400 0.9278
0.1305 0.42 440 0.2404 0.9232
0.2595 0.43 450 0.2389 0.9209
0.1901 0.44 460 0.2102 0.9266
0.1993 0.45 470 0.2129 0.9255
0.147 0.46 480 0.2208 0.9232
0.1801 0.47 490 0.2143 0.9255
0.1716 0.48 500 0.2416 0.9209
0.1281 0.48 510 0.2152 0.9232
0.1837 0.49 520 0.2112 0.9243
0.1681 0.5 530 0.2178 0.9232
0.1408 0.51 540 0.2127 0.9243
0.1229 0.52 550 0.3322 0.9278
0.1304 0.53 560 0.3586 0.9209
0.1905 0.54 570 0.3354 0.9243
0.147 0.55 580 0.3431 0.9278
0.1538 0.56 590 0.3444 0.9232
0.1504 0.57 600 0.2196 0.9266
0.1628 0.58 610 0.3452 0.9163
0.1387 0.59 620 0.3282 0.9278
0.2104 0.6 630 0.2132 0.9243
0.1482 0.61 640 0.2154 0.9243
0.217 0.62 650 0.3472 0.9197
0.1692 0.63 660 0.2063 0.9243
0.175 0.64 670 0.2019 0.9278
0.1473 0.65 680 0.1957 0.9266
0.1154 0.66 690 0.2020 0.9255
0.1369 0.67 700 0.2087 0.9266
0.1262 0.67 710 0.3224 0.9289
0.2111 0.68 720 0.3325 0.9243
0.1349 0.69 730 0.3285 0.9289
0.1814 0.7 740 0.3324 0.9266
0.1217 0.71 750 0.3212 0.9243
0.173 0.72 760 0.2176 0.9220
0.1441 0.73 770 0.2130 0.9232
0.1706 0.74 780 0.2136 0.9220
0.1411 0.75 790 0.2101 0.9220
0.1051 0.76 800 0.2078 0.9243
0.115 0.77 810 0.2160 0.9266
0.2031 0.78 820 0.2162 0.9209
0.12 0.79 830 0.2059 0.9255
0.176 0.8 840 0.2100 0.9255
0.1306 0.81 850 0.4307 0.9243
0.1359 0.82 860 0.4397 0.9289
0.1921 0.83 870 0.5446 0.9278
0.1772 0.84 880 0.5423 0.9266
0.1771 0.85 890 0.4273 0.9266
0.1965 0.86 900 0.3224 0.9243
0.1227 0.86 910 0.2131 0.9278
0.2046 0.87 920 0.3130 0.9278
0.1061 0.88 930 0.3180 0.9289
0.1364 0.89 940 0.5501 0.9186
0.1213 0.9 950 0.4400 0.9220
0.1611 0.91 960 0.4364 0.9255
0.1632 0.92 970 0.4475 0.9220
0.1617 0.93 980 0.5758 0.9209
0.1478 0.94 990 0.2143 0.9220
0.1314 0.95 1000 0.2156 0.9232
0.1814 0.96 1010 0.2191 0.9220
0.1669 0.97 1020 0.2129 0.9243
0.1206 0.98 1030 0.2119 0.9220
0.1852 0.99 1040 0.2104 0.9209
0.1381 1.0 1050 0.1999 0.9255
0.135 1.01 1060 0.2090 0.9243
0.1253 1.02 1070 0.4486 0.9209
0.1244 1.03 1080 0.4319 0.9197
0.1772 1.04 1090 0.4248 0.9243
0.1264 1.05 1100 0.3090 0.9289
0.6928 1.05 1110 0.3174 0.9278
0.0908 1.06 1120 0.4359 0.9266
0.1286 1.07 1130 0.4302 0.9312
0.0953 1.08 1140 0.5397 0.9289
0.1091 1.09 1150 0.5455 0.9255
0.1546 1.1 1160 0.4261 0.9300

Framework versions

  • Transformers 4.34.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.5
  • Tokenizers 0.14.1
Downloads last month
13
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for vxbrandon/t5-base_sst2_dense

Base model

google-t5/t5-base
Finetuned
(430)
this model

Dataset used to train vxbrandon/t5-base_sst2_dense

Evaluation results