CoolFace
Modelpublic

paulo037/20260503T212014Z-conll03_ner-graft-soft-invalid-b00-4b-10k-3d86cde4

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face
0likes7downloads
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. -->

20260503T212014Z-conll03_ner-graft-soft-invalid-b00-4b-10k-3d86cde4

This model is a fine-tuned version of Qwen/Qwen3-VL-4B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.0029
  • —Graft/constrained Ratio: 1.0
  • —Graft/constrained Tokens: 10821.0
  • —Graft/fallback Samples: 0.0
  • —Graft/standard Fallback Samples: 0.0

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: 0.0001
  • —trainbatchsize: 1
  • —evalbatchsize: 1
  • —seed: 14
  • —gradientaccumulationsteps: 128
  • —totaltrainbatch_size: 128
  • —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_ratio: 0.03
  • —num_epochs: 3

Training results

Training LossEpochStepValidation LossRatioTokensSamplesFallback Samples
0.02530.3328260.01081.022433.00.00.0
0.01210.6656520.00661.017626.00.00.0
0.00750.9984780.00471.010821.00.00.0
0.00851.321040.00361.023581.00.00.0
0.0051.65281300.00371.017229.00.00.0
0.00551.98561560.00451.010821.00.00.0
0.00242.30721820.00321.023207.00.00.0
0.00252.642080.00331.016213.00.00.0
0.00212.97282340.00291.010821.00.00.0

Framework versions

  • —PEFT 0.17.1
  • —Transformers 4.57.0
  • —Pytorch 2.11.0a0+eb65b36914.nv26.02
  • —Datasets 4.1.1
  • —Tokenizers 0.22.2