CoolFace
Modelpublic

osmajic/paligemma_scienceQA_v15

sourceHugging Facegemmaupdated 2y agoView on Hugging Face
0likes4downloads
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. -->

paligemmascienceQAv15

This model is a fine-tuned version of google/paligemma-3b-pt-224 on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.1043

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: 2e-06
  • —trainbatchsize: 2
  • —evalbatchsize: 2
  • —seed: 42
  • —gradientaccumulationsteps: 4
  • —totaltrainbatch_size: 8
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 5

Training results

Training LossEpochStepValidation Loss
0.15020.2100.1155
0.14140.4200.1136
0.14810.6300.1134
0.13840.8400.1121
0.14721.0500.1112
0.13571.2600.1108
0.1241.4700.1102
0.14951.6800.1096
0.14381.8900.1090
0.13942.01000.1078
0.12852.21100.1071
0.1312.41200.1071
0.13752.61300.1068
0.12432.81400.1061
0.1383.01500.1056
0.13863.21600.1055
0.13253.41700.1052
0.12523.61800.1052
0.13713.81900.1049
0.1224.02000.1048
0.11494.22100.1046
0.13184.42200.1044
0.11694.62300.1043
0.14124.82400.1041
0.15145.02500.1043

Framework versions

  • —PEFT 0.12.0
  • —Transformers 4.43.3
  • —Pytorch 2.3.1+cu121
  • —Datasets 2.20.0
  • —Tokenizers 0.19.1