CoolFace
Modelpublic

gaodrew/latin_gpt2

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes19downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

latin_gpt2

This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset. It achieves the following results on the evaluation set:

  • —Loss: 5.4614

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: 0.0005
  • —trainbatchsize: 512
  • —evalbatchsize: 512
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_steps: 3000
  • —num_epochs: 1
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation Loss
7.1720.00575007.1975
6.72230.011510006.9021
6.46260.017215006.6753
6.27430.023020006.5041
6.16690.028725006.3793
6.12120.034430006.3379
6.05110.040235006.3103
5.99190.045940006.2384
5.93640.051645006.2122
5.8870.057450006.1205
5.85640.063155006.1243
5.82640.068960006.0895
5.79420.074665006.0713
5.78260.080370006.0359
5.74750.086175006.0351
5.73290.091880006.0039
5.71270.097585005.9915
5.69710.103390005.9835
5.68250.109095005.9810
5.66950.1148100005.9542
5.65370.1205105005.9367
5.64550.1262110005.9178
5.62540.1320115005.9097
5.620.1377120005.9092
5.6080.1434125005.8881
5.59930.1492130005.8807
5.58910.1549135005.8707
5.57910.1607140005.8809
5.57010.1664145005.8585
5.56510.1721150005.8436
5.55950.1779155005.8607
5.55090.1836160005.8308
5.54010.1894165005.8381
5.5350.1951170005.8749
5.52810.2008175005.8331
5.52310.2066180005.8139
5.51480.2123185005.8078
5.51120.2180190005.8016
5.50490.2238195005.8034
5.50060.2295200005.8025
5.49090.2353205005.8017
5.48350.2410210005.7782
5.48410.2467215005.7862
5.47940.2525220005.7690
5.4760.2582225005.7689
5.46680.2639230005.7806
5.45850.2697235005.7678
5.45730.2754240005.7499
5.45510.2812245005.7696
5.4510.2869250005.7564
5.44650.2926255005.7508
5.43960.2984260005.7414
5.43560.3041265005.7354
5.43210.3098270005.7471
5.4270.3156275005.7296
5.42420.3213280005.7294
5.41920.3271285005.7252
5.41680.3328290005.7183
5.41350.3385295005.7241
5.40770.3443300005.7148
5.40510.3500305005.7215
5.39940.3557310005.7140
5.39920.3615315005.7079
5.39020.3672320005.7057
5.38480.3730325005.7047
5.38650.3787330005.6973
5.38240.3844335005.6938
5.37690.3902340005.6950
5.37330.3959345005.6885
5.36940.4017350005.6819
5.36380.4074355005.6770
5.36110.4131360005.6819
5.36150.4189365005.6705
5.3540.4246370005.6757
5.35220.4303375005.6718
5.34220.4361380005.6679
5.3430.4418385005.6655
5.34340.4476390005.6591
5.33850.4533395005.6608
5.33250.4590400005.6629
5.33150.4648405005.6581
5.33170.4705410005.6534
5.32750.4762415005.6447
5.32020.4820420005.6451
5.31490.4877425005.6348
5.3130.4935430005.6366
5.31220.4992435005.6384
5.30650.5049440005.6326
5.2990.5107445005.6226
5.29970.5164450005.6301
5.29590.5221455005.6172
5.29070.5279460005.6232
5.28890.5336465005.6239
5.2880.5394470005.6069
5.2770.5451475005.6154
5.27650.5508480005.6157
5.27390.5566485005.6035
5.26930.5623490005.6009
5.26350.5681495005.5978
5.26820.5738500005.5987
5.2580.5795505005.5971
5.260.5853510005.5994
5.25680.5910515005.5873
5.24460.5967520005.5771
5.24690.6025525005.5824
5.24590.6082530005.5853
5.2390.6140535005.5781
5.23550.6197540005.5729
5.22960.6254545005.5737
5.23010.6312550005.5656
5.22770.6369555005.5716
5.21970.6426560005.5583
5.21310.6484565005.5639
5.21320.6541570005.5537
5.21030.6599575005.5656
5.20780.6656580005.5524
5.2030.6713585005.5470
5.20350.6771590005.5454
5.19630.6828595005.5428
5.19320.6885600005.5355
5.19060.6943605005.5338
5.18640.7000610005.5352
5.18230.7058615005.5295
5.1790.7115620005.5296
5.17520.7172625005.5259
5.17260.7230630005.5291
5.16920.7287635005.5183
5.16940.7345640005.5173
5.16240.7402645005.5162
5.1610.7459650005.5104
5.15880.7517655005.5145
5.1560.7574660005.5057
5.15390.7631665005.5036
5.14740.7689670005.5093
5.14570.7746675005.5052
5.14440.7804680005.4979
5.14370.7861685005.4979
5.14020.7918690005.5009
5.13460.7976695005.4907
5.13080.8033700005.4905
5.13250.8090705005.4880
5.12970.8148710005.4866
5.12460.8205715005.4871
5.12320.8263720005.4846
5.12320.8320725005.4840
5.12190.8377730005.4811
5.11260.8435735005.4791
5.11680.8492740005.4782
5.11750.8549745005.4763
5.11080.8607750005.4771
5.10820.8664755005.4742
5.10530.8722760005.4738
5.1070.8779765005.4718
5.10740.8836770005.4694
5.10740.8894775005.4692
5.10380.8951780005.4697
5.1050.9008785005.4684
5.10190.9066790005.4665
5.10150.9123795005.4663
5.1010.9181800005.4672
5.10220.9238805005.4654
5.10.9295810005.4632
5.09810.9353815005.4637
5.09810.9410820005.4630
5.09510.9468825005.4619
5.09410.9525830005.4628
5.09470.9582835005.4621
5.09490.9640840005.4625
5.09710.9697845005.4618
5.08950.9754850005.4619
5.09370.9812855005.4616
5.09780.9869860005.4615
5.09580.9927865005.4614
5.0950.9984870005.4614

Framework versions

  • —Transformers 4.41.2
  • —Pytorch 2.2.0
  • —Datasets 2.20.0
  • —Tokenizers 0.19.1